prompt_id,source_run,source_iteration,question,ddc_domain,ddc_category,difficulty,has_context,register,intended_skill,topic
q-corpus-1/q100_1/1,q100_1,1,"I'm building a lock-free concurrent queue using a Michael-Scott design in C++20, but I'm hitting a wall with the ABA problem during the `compare_exchange_weak` on the tail pointer. 

Here is the current state of my `dequeue` logic:
- I'm using `std::atomic<Node*>` for head and tail.
- Nodes are dynamically allocated and freed via `delete`.
- My current attempt at hazard pointers is causing a deadlock during the reclamation phase because the retirement list is growing faster than the scan interval.

Can you rewrite my `dequeue` implementation to resolve the ABA problem without using hazard pointers? Specifically, implement it using a tagged pointer approach (double-wide CAS) if the architecture supports it, or provide a version using `std::atomic<std::shared_ptr<Node>>` that avoids the performance overhead of traditional reference counting. Please include the logic for the head-tail synchronization to ensure that a slow producer doesn't leave the tail pointing to a reclaimed node.","0 Computer science, information, and general works","00 Computer science, knowledge, and systems",a very difficult,False,human_casual,coding,"0 Computer science, information, and general works"
q-corpus-1/q100_1/2,q100_1,2,"A 68-year-old patient in the ICU has hypotension and a pulmonary artery catheter in place. Severe tricuspid regurgitation is suspected, and I need to compare two named cardiac output measurement methods using the same bedside data.

Calculate cardiac output by the direct Fick method and by thermodilution, then explain the key differences between the Fick method and thermodilution that would make their results diverge in this clinical situation.

Data for Fick calculation:

- Oxygen consumption, VO₂ = 250 mL/min
- Hemoglobin = 10.0 g/dL
- Arterial oxygen saturation = 98%
- Mixed venous oxygen saturation = 60%
- PaO₂ = 100 mmHg
- PvO₂ = 30 mmHg
- Oxygen-binding capacity of hemoglobin = 1.34 mL O₂/g Hb
- Dissolved oxygen coefficient = 0.003 mL O₂/dL/mmHg

Data for thermodilution calculation:

- Injectate volume = 10 mL
- Blood temperature = 37°C
- Injectate temperature = 0°C
- Computation constant K = 0.825
- Area under the thermal dilution curve = 8.0 °C·s

Use the Fick equation:

\[
CO = \frac{VO_2}{CaO_2 - CvO_2}
\]

where arterial and venous oxygen content include both hemoglobin-bound and dissolved oxygen.

For thermodilution, use:

\[
CO = \frac{V_i \times (T_b - T_i) \times K}{AUC}
\]

and convert the result to L/min.

Provide one response containing:

1. Cardiac output by the Fick method, in L/min.
2. Cardiac output by thermodilution, in L/min.
3. A concise quantitative or formal explanation of the key differences between these two named methods that best accounts for any discrepancy, especially in the presence of tricuspid regurgitation.",6 Technology,61 Medicine and health,a,False,human_professional,math_reasoning,6 Technology
q-corpus-1/q100_1/3,q100_1,3,"1. US CPI inflation was 1.6% in 1965 with unemployment at 4.5%, and 5.5% in 1970 with unemployment at 4.9%.
2. Following the 1973 oil price shock, CPI inflation reached 11.0% in 1974 and unemployment rose to 5.6% in 1975.
3. Average CPI inflation was 10.9% from 1978-1980 with unemployment averaging 6.8% over 1979-1980.
4. 
```
Year  CPI YoY  Unemployment
1965   1.6%      4.5%
1970   5.5%      4.9%
1974  11.0%      5.6%
1979   13.3%      5.8%
1980   12.5%      7.1%
```
5. Friedman 1968, “The Role of Monetary Policy”, states: “Inflation is always and everywhere a monetary phenomenon in the sense that it is and can be produced only by a more rapid increase in the quantity of money than in output.”
6. Phelps 1967 and Friedman 1968 introduced the natural rate of unemployment and argued that systematic policy attempts to exploit a short-run Phillips trade-off would be undone by expectations adjustment.

Explain how the price and labor market outcomes in 1965-1980 are invoked to support the natural rate hypothesis and the monetarist critique of the Phillips curve that informed the shift toward rules-based monetary policy in the early 1980s.",3 Social Science,33 Economics,a difficult,False,human_professional,world_knowledge,3 Social Science
q-corpus-1/q100_1/4,q100_1,4,"Hey, we just rolled out Digital Collections Access Policy v2.3 effective 2025-01-01 and the access checker is still on the old logic. We got two complaints this week about theses showing as open too early and a 1932 newspaper showing as open off-campus.

Policy excerpt v2.3:
1. Theses: author can set an embargo_end date. Max initial embargo is 5 years from degree date, renewable once for another 5 years. Total embargo may not exceed 10 years. After embargo_end the item must be open unless item['withdrawn'] is true, in which case access is 'withdrawn'. If embargo_end is missing or in the past, treat as open unless withdrawn.
2. Newspaper issues: public domain if publication_year <= current_year - 95. Otherwise access is 'campus' for on-campus IPs and 'restricted' off-campus. Authenticated users always get 'campus' for non-public domain.
3. Special collections: access level is always 'restricted' unless item['dean_override'] is true, in which case access is 'open'.

Current implementation in access_check.py:
def check_access(item, today):
    typ = item.get('type')
    if typ == 'thesis':
        embargo_end = item.get('embargo_end')
        if embargo_end and embargo_end > today:
            return 'restricted'
        return 'open'
    elif typ == 'newspaper':
        year = item.get('publication_year')
        if year and year < today.year - 95:
            return 'open'
        if item.get('on_campus'):
            return 'campus'
        return 'restricted'
    elif typ == 'special':
        if item.get('dean_override'):
            return 'open'
        return 'restricted'
    return 'open'

Items use ISO date strings for embargo_end, today is a date object, and booleans for on_campus, authenticated, dean_override, withdrawn.

Can you rewrite check_access so it implements the policy exactly for theses, newspapers and special collections, including the 10-year max embargo cap, withdrawn handling, and the correct public-domain cutoff and authenticated user rule? Return the corrected Python function.","0 Computer science, information, and general works",02 Library and information sciences,a difficult,True,human_casual,coding,"0 Computer science, information, and general works"
q-corpus-1/q100_1/5,q100_1,5,"Hey, I’m revisiting the English that-trace effect and got stuck on the adverb-rescue facts. The standard pattern is sharp:

 1. *Who do you think that ___ left?
 2. Who do you think ___ left?
 3. Who do you think that John saw?

 But an intervening adverb salvages the exact configuration that is otherwise blocked:

 4. Who do you think that probably ___ left?

 What specific syntactic or parsing-based analysis has been proposed to explain why an adverb like “probably” licenses subject extraction in the presence of an overt complementizer, and how does this bear on whether the phenomenon is best derived from the ECP or from a linear adjacency condition?",4 Language,40 Language,a difficult,False,human_casual,world_knowledge,4 Language
q-corpus-1/q100_1/6,q100_1,6,"1. A helical compression spring for an engine valve must provide a rate of `k = 10.0 N/mm` and sustain a maximum working load `F = 80 N` without yielding. The spring must fit inside an outer‑diameter envelope of 25 mm, so that `D + d ≤ 25 mm`, where `D` is the mean coil diameter and `d` is the wire diameter.

2. The material is chromium‑vanadium steel with shear modulus `G = 79 GPa`, density `ρ = 7.80×10⁻⁶ kg/mm³`, and allowable shear stress `τ_a = 350 MPa`. The stress equation uses the Wahl factor:
   `τ = 8·F·D·K / (π·d³)`,  
   where `K = (4C−1)/(4C−4) + 0.615/C` and `C = D/d`.  
   The design index must satisfy `4 ≤ C ≤ 12`.

3. The spring rate is analytically: `k = G·d⁴ / (8·D³·N)`, where `N` is the number of active coils.

4. The spring’s first natural frequency is computed for a fixed‑free condition with end mass neglected but effective inertia approximated as `m_eff = m/3`, where `m` is the total wire mass. The mass is:
   `m = ρ · (πd²/4) · (π·D·N)`.  
   The valve imposes the requirement `f = (1/2π)·√(k/ m_eff ) ≥ 80 Hz`, which translates to `m ≤ 3k / (2π·80)²` (with `k` in N/mm and `m` in kg – keep consistent units).

5. The spring must not fail by buckling or solid contact, but these are not limiting for the proposed geometry; only the above constraints are to be respected.

All material data, formulas, and limits are given above.

Determine the wire diameter `d`, mean coil diameter `D`, number of active coils `N`, and the resulting total spring mass `m` that meets the valve’s load, rate, fatigue stress, index limit, and frequency requirement. Identify the governing (active) constraint. Show your calculation in full.",6 Technology,62 Engineering,a very difficult,True,agent_structured,math_reasoning,6 Technology
q-corpus-1/q100_1/7,q100_1,7,"Claim: Implementing a permanent 100% tax on capital gains for high-net-worth individuals will eliminate the ""lock-in effect"" and increase overall market liquidity without reducing long-term capital investment.

Inputs:
- Lock-in effect definition: The tendency of investors to hold onto appreciated assets to avoid realizing capital gains taxes.
- Current baseline: A capital gains tax rate of 20%.
- Market assumption: Investors operate on a discounted present value basis and are sensitive to the after-tax internal rate of return (IRR).

Task: Provide a rigorous economic analysis of the arguments for and against this claim. Specifically, address the trade-off between the removal of the lock-in effect and the potential distortion of investment incentives (deadweight loss). Analyze whether a 100% rate fundamentally alters the asset-holding decision compared to a high but non-prohibitive rate (e.g., 50%), and explain how this affects the allocation of capital across different risk profiles.",3 Social Science,33 Economics,a difficult,False,agent_structured,world_knowledge,3 Social Science
q-corpus-1/q100_1/8,q100_1,8,"I have a chunk of obsidian on my desk that looks exactly like black glass, with no visible mineral grains like I'd expect from a normal rock. What exactly is obsidian, and how does it form in the rock cycle to end up looking like that?",5 Science,55 Earth sciences and geology,a very simple,False,human_casual,world_knowledge,5 Science
q-corpus-1/q100_1/9,q100_1,9,"Consider the invented language Lê with verb root `paku` 'wash'. Two productive suffixes attach to the root: `-chi` (causative) and `-ri` (reciprocal). Native judgments for forms and available interpretations are:

1. `paku-chi` - 'cause to wash' - grammatical
2. `paku-ri` - 'wash each other' - grammatical
3. `paku-chi-ri` - grammatical ONLY with reading A: 'X causes Y and Z to wash each other' (X is causer, Y/Z are reciprocal agents/patients). Reading B: 'Y and Z each cause someone (else) to wash' is impossible for this form, and judged ungrammatical if forced.
4. `paku-ri-chi` - grammatical ONLY with reading B: 'Y and Z each cause someone to wash'. Reading A is impossible for this form.
5. `paku-chi-ri-chi` - ungrammatical on any reading.
6. Speakers reject any form where `-ri` scopes over `-chi` but `-chi` is linearly outermost, and vice versa.

Assume for your argument:
- Syntax is built by Merge. Causative head `Chi` introduces a causer argument and takes a VP complement; reciprocal head `Ri` introduces reciprocal semantics and also takes a VP complement. One must asymmetrically c-command the other.
- Morphology is derived by successive head-movement of V to the next head above it, left-adjoining, obeying the Mirror Principle: linear order AFF1-AFF2 on the verb directly reflects hierarchical order [ AFF2 [ AFF1 [ VP ] ] ] (the affix closer to the root corresponds to the lower head).
- A competing templatic hypothesis H-Temp states: Lê has a fixed morphological template Root-chi-ri regardless of scope, and interpretation is determined separately/pragmatically, not by hierarchical order.

Prove that H-Temp cannot derive the pattern in (3)-(4) without positing either accidental homophony or an extra stipulated zero morpheme, and derive the unique hierarchical structure [ Ri [ Chi [ VP ] ] ] vs [ Chi [ Ri [ VP ] ] ] that under H-Mirror predicts exactly the attested correlation between linear order and interpretation in (3)-(4) while correctly ruling out (5) and the missing readings. Your proof must give explicit tree derivations for both attested forms via head-movement, show why the alternative hierarchy fails to generate one of the readings, and show by exhaustive enumeration of possible Merge orders under H-Temp why it overgenerates the unattested reading and undergenerates nothing without additional stipulation.",4 Language,40 Language,a very difficult,False,agent_structured,math_reasoning,4 Language
q-corpus-1/q100_1/10,q100_1,10,"Glenn Gould recorded Bach's Goldberg Variations (BWV 988) twice — in 1955 (Columbia ML 5060, New York, age 22) and in 1981 (CBS IM 37779, Toronto, age 50, weeks before his death). The 1955 version became a landmark debut and one of the best-selling classical piano recordings ever made; it is fleet, light, and rhythmically crisp. The 1981 version is markedly slower, with weighted bass, deeper voicing, and longer silences between variations. Gould called the later recording the one he'd ""rather not have to live without.""

What is the principal interpretive shift between Gould's 1955 and 1981 Goldberg Variations, and what does the contrast reveal about how a single performer's conception of one work can change across a career?",7 Arts & recreation,78 Music,a simple,True,agent_structured,world_knowledge,7 Arts & recreation
q-corpus-1/q100_1/11,q100_1,11,"I'm finalizing the wall captions for our French Revolution gallery before the printer's deadline this afternoon. Can you confirm the name of the medieval fortress-prison in Paris that revolutionaries stormed on July 14, 1789?",9 History,90 History,a very simple,False,human_casual,world_knowledge,9 History
q-corpus-1/q100_1/12,q100_1,12,"hey, what are the key differences between Mario Kart 8's luck-driven item drops and Splatoon 2's skill-based ink gun aiming? could you rewrite the start-of-race call from Mario Kart 8's Royal Raceway so it's shouted by a Splatoon 2 announcer pumping up turf coverage instead of lap counts? original line: ""Welcome to Royal Raceway, racers! Grab a shell or mushroom and take the corners—first across the line takes the cup!"" give me the rewritten bite and a two-sentence contrast on why the two games pull players differently.",7 Arts & recreation,"79 Sports, games and entertainment",a,False,human_casual,creative_synthesis,7 Arts & recreation
q-corpus-1/q100_1/13,q100_1,13,"Available tools:
- update_ticket_status(ticket_id: str [required], status: str, enum: ['open','in_progress','waiting','resolved','closed'] [required], comment: str = '') -> bool
  Update the status of a support ticket.
- calculate_invoice(line_items: list[dict] [required], tax_rate: float = 0.0, discount_percent: float = 0.0) -> dict with subtotal, tax, discount, total
  Calculate the total for an invoice including tax and discounts.
- fuzzy_match(query: str [required], dataset: str [required], threshold: float = 0.8, max_results: int = 5) -> list of match dicts with value and score
  Find closest matches to a query string in a dataset.
- semantic_search(query: str [required], collection: str [required], top_k: int = 10) -> list of result dicts with id, text, similarity_score
  Perform semantic vector search on a document collection.

Known issues:
- Warning: update_ticket_status can timeout after 30 seconds if the underlying service is under heavy load. Plan for retries.

Ticket T-7742 needs to be resolved with a $420 credit applied for the SLA breach on 2024-11-15. First, search the resolution_guides collection for ""SLA breach credit calculation"" to confirm the standard credit amount and required documentation. Then calculate the invoice line item for the credit (single line: ""SLA breach credit - Nov 15"", quantity 1, unit price -420) with 0% tax and 0% discount. Only after both steps succeed, update T-7742 to ""resolved"" with comment ""Credit applied per resolution guide R-2201"". The ticketing service has been flaky — if the status update times out, retry once after 45 seconds but do not re-run the search or calculation.",A Agentic tool use,single_call_error_anticipation,a difficult,True,agent_structured,planning_agentic,A Agentic tool use
q-corpus-1/q100_1/14,q100_1,14,"I'm putting together program notes for a board game night themed around ""the evolution of worker placement,"" and I want to open with a short comparative writeup on Agricola (2007) and Caverna: The Cave Farmers (2013), both by Uwe Rosenberg.

Could you write a comparison covering these points:

1. Core resource-and-feeding loop: how the harvest/feeding mechanism differs between the two (Agricola's harvest phase penalties vs. Caverna's approach), and what effect that has on how punishing early mistakes are.
2. The role of the occupation/minor improvement card decks in Agricola versus their removal in Caverna (replaced by fixed player boards with built-in special abilities) — what this does to variability between plays and to the game's dependence on card-drafting or card-luck.
3. Expansion of the action space: Agricola's action board grows modestly with player count, while Caverna's is considerably larger and includes a dedicated combat/adventure sub-track. Explain how this changes the decision density per turn and the viability of divergent strategies (e.g., pure farming vs. mining/ruby-rush vs. weapons-and-adventuring).
4. Overall pacing and playtime differences across a standard 2-player game, and which design is generally considered friendlier to new players versus experienced optimizers.

Please close with a short verdict on which title better rewards long-term mastery and which is more approachable for a first-time worker-placement audience, citing specific mechanical reasons from the points above rather than general taste.",7 Arts & recreation,"79 Sports, games and entertainment",a,False,human_professional,world_knowledge,7 Arts & recreation
q-corpus-1/q100_1/15,q100_1,15,"Compare Bronisław Malinowski's ethnographic account of the Kula ring among the Trobriand Islanders with Marcel Mauss's theory of the gift as articulated in The Gift (1925). How does Mauss define the gift as a 'total social fact,' and which single concrete detail from Malinowski's Kula description either fulfills or problematizes that definition?",3 Social Science,"30 Social sciences, sociology, and anthropology",a,False,agent_structured,world_knowledge,3 Social Science
q-corpus-1/q100_1/16,q100_1,16,"Compare the Meissner effect in a Type-I superconductor with what happens in a Type-II superconductor once the applied field exceeds Hc1.

Specifically:

1. Explain why a Type-I superconductor below Hc expels magnetic field lines entirely, and connect this to the London penetration depth and the condition of zero electrical resistance combined with perfect diamagnetism (why resistance-free current alone doesn't guarantee flux expulsion, but the Meissner state requires more).

2. For a Type-II superconductor between Hc1 and Hc2, describe the vortex (mixed) state: how flux enters as quantized fluxoids, each carrying one flux quantum Φ0 = h/2e, and how the vortex core size relates to the coherence length ξ versus the London penetration depth λ.

3. Using the Ginzburg-Landau parameter κ = λ/ξ, explain the criterion (κ > 1/√2) that distinguishes Type-II from Type-I behavior, and why this makes Type-II materials suitable for high-field magnets (e.g., Nb-Ti or Nb3Sn) while Type-I superconductors (e.g., lead, mercury) are not.

Conclude with a short discussion of what happens physically to the vortex lattice as flux pinning is introduced (e.g., via lattice defects) and why pinning is essential for maintaining zero resistive loss in practical high-current applications.",5 Science,53 Physics,a,False,agent_structured,world_knowledge,5 Science
q-corpus-1/q100_1/17,q100_1,17,"Available tools:
- send_sms(phone_number: str [required], message: str [required]) -> dict with message_id and status
  Send an SMS text message.

Our chat bridge is down, so SMS is the fallback. I need the exact two-step send_sms plan for a fact verification and trace.

1. Step 1: call send_sms for phone_number +15550142299 with a message in this exact shape:
   ""Fact: <answer>""
   where <answer> is the capital of France.

2. Step 2: call send_sms for phone_number +15550142300 with a message in this exact shape:
   ""Trace: <message_id>/<status>""
   substituting the message_id and status returned by Step 1.

Outcome: a verified fact notification and a dependent trace confirmation. Provide the constructed phone_number and message arguments for both calls, and state how Step 2 depends on Step 1.",A Agentic tool use,parameter_construction,a very simple,True,human_professional,planning_agentic,A Agentic tool use
q-corpus-1/q100_1/18,q100_1,18,Explain the symbolism of the bull and the horse in Pablo Picasso's Guernica.,7 Arts & recreation,70 Arts,a simple,False,human_casual,world_knowledge,7 Arts & recreation
q-corpus-1/q100_1/19,q100_1,19,"Section 7.3.2 of specification PCB-CTRL-2024 Rev C states: ""The dual-clock status FIFO bridges the 48 MHz sensor clock and 12 MHz bus clock using gray-coded read/write pointers. Two-stage synchronizer on pointer bus mitigates metastability. Constraint SYS-14 mandates ≥350 ps synchronizer slack at minimum operating voltage."" Dr. L's review comment asserts: ""Expert claim: power-aware gate-level simulation (tool PowSim) must run before formal equivalence checking (tool EquiCheck) on the pointer synchronizer; if EquiCheck runs first, clock-gating inserts from synthesis hide metastability protocol breaches that formal models label unreachable."" Internal toolchain notes: PowSim ingests placed netlist and pow.ann, emits trace.pds. EquiCheck requires RTL and netlist; accepts trace.pds to restrict initial states and inject observed gating patterns.

I’m finalizing the verification flow for the rev C pump controller FIFO and need your input: lay out the dependency-ordered sequence that calls PowSim then EquiCheck on this synchronizer, specifying exactly how the trace.pds output must feed EquiCheck’s constrained initial states and gating injection, and within that same plan present the strongest arguments for and against Dr. L’s claim that PowSim must precede EquiCheck to avoid masked metastability faults under SYS-14. Work only from the netlist, RTL, and pow.ann referenced above.",6 Technology,62 Engineering,a very difficult,True,human_professional,planning_agentic,6 Technology
q-corpus-1/q100_1/20,q100_1,20,"We are standardizing an archival pedigree chart for Augusta Ada King, Countess of Lovelace (1815–1852) using the Sosa–Stradonitz (Ahnentafel) numbering system. 

The initial nodes of the chart are defined as follows:

| Ahnentafel Number | Individual | Relationship to Proband |
| :--- | :--- | :--- |
| 1 | Ada Lovelace | Proband (Subject) |
| 2 | George Gordon Byron, 6th Baron Byron | Father |
| 3 | Anne Isabella Milbanke | Mother |
| 5 | Catherine Gordon of Gight | Paternal Grandmother |

Under standard Ahnentafel rules, what is the exact index assigned to Catherine Gordon's father (George Gordon, 12th Laird of Gight), and what formula determines this position relative to his daughter?",9 History,92 Biography and genealogy,a simple,False,human_professional,world_knowledge,9 History
q-corpus-1/q100_1/21,q100_1,21,"Remind me of the basic distinction between statutory incidence and economic incidence? Comms is drafting a note that says our new $0.02-per-ounce sugary-drink tax is “paid by distributors,” and I don’t want that wording if economists mean something else by who the tax actually falls on.",3 Social Science,33 Economics,a very simple,False,human_casual,world_knowledge,3 Social Science
q-corpus-1/q100_1/22,q100_1,22,"Explain the phonological rule governing English's ""a"" versus ""an"", and show how it handles words like ""an hour"", ""a university"", and ""an honest man"" — what phonetic property actually decides the alternation, and can orthography ever force the opposite choice (as some older prescriptions attempt with ""an historic"")?",4 Language,40 Language,a very simple,False,agent_structured,world_knowledge,4 Language
q-corpus-1/q100_1/23,q100_1,23,"1. Text: Bernard Williams, “Integrity,” in J. J. C. Smart and Bernard Williams, Utilitarianism: For and Against (Cambridge University Press, 1973), especially the Jim and the Indians and George the chemist episodes.
2. Situation: A public-health statistician must decide whether to sign a mortality model that slightly smooths uncertain nursing-home deaths so a legislature will release emergency funding; if she does not sign, the funding is likely to be delayed until after winter.
3. Constraint: Ground the answer in Williams’s objection to utilitarianism through negative responsibility and the idea that an agent cannot simply be a conduit for optimal outcomes.
4. Constraint: Exclude rights-based, legal, professional-code, and virtue-ethical vocabularies.
5. Constraint: Address what is at stake in her being asked to make the smoothing her own, not whether the legislature should receive funding.

Explain what Williams’s integrity objection shows to be ethically at stake in signing the smoothed model.",1 Philosophy and psychology,17 Ethics,a very difficult,False,agent_structured,world_knowledge,1 Philosophy and psychology
q-corpus-1/q100_1/24,q100_1,24,"{""request_id"":""hist-004"",""operation"":""recall"",""subject"":""Meiji Restoration"",""prompt"":""In 1868, a teenage emperor was restored as the nominal ruler of Japan after the fall of the Tokugawa shogunate. Provide only his era name.""}",9 History,90 History,a very simple,False,agent_structured,world_knowledge,9 History
q-corpus-1/q100_1/25,q100_1,25,I'm documenting a question from our intake team: a client recently informed their therapist of plans to drive while severely intoxicated. The therapist is uncertain whether they may legally and ethically notify the client's employer or family without violating confidentiality. What is the name of the APA Ethical Principles standard that specifically permits breaching confidentiality to protect others from foreseeable harm?,1 Philosophy and psychology,17 Ethics,a simple,False,human_professional,world_knowledge,1 Philosophy and psychology
q-corpus-1/q100_1/26,q100_1,26,"Input: An undirected weighted graph G with vertex set V={0,…,n−1} (n even) and edge set E composed of: (i,i+1 mod n) for all i (the n‑cycle), plus chords (i,i+n/2 mod n) for i=0,…,n/2−1. All edge weights are distinct positive integers ≤ W (the exact values are irrelevant for the count). Source vertex s=0.

Task: Run Dijkstra’s algorithm using a binary min‑heap. Compute the exact number of extract‑min operations and the exact number of edge relaxations performed on G, expressed as a function of n. Prove that the total running time is Θ(n log n) for this family of graphs.

Counterfactual: If the chords are removed (so G becomes the n‑cycle, which is a tree after deleting any one cycle edge), what are the new counts of extract‑min and relaxation steps, and what asymptotic running time does Dijkstra achieve on that reduced graph? Provide the revised counts and a concise proof.","0 Computer science, information, and general works","00 Computer science, knowledge, and systems",a difficult,False,agent_structured,math_reasoning,"0 Computer science, information, and general works"
q-corpus-1/q100_1/27,q100_1,27,"Using USGS gauge data for the Mississippi River at Vicksburg, MS (station 07289000), where the long-term mean annual discharge is approximately 566,000 cfs and the April 2011 flood peak discharge was approximately 2,310,000 cfs, calculate the percent by which the flood peak exceeded the mean annual discharge.",5 Science,55 Earth sciences and geology,a very simple,False,agent_structured,math_reasoning,5 Science
q-corpus-1/q100_1/28,q100_1,28,"1. Rio de Janeiro lies at 22.9°S, 43.2°W on a sheltered Atlantic bay, while Brasília lies at 15.8°S, 47.9°W on the Brazilian plateau, roughly 1,100 km inland.
2. Rio’s humid tropical climate averages about 1,200 mm of annual rainfall; Brasília has a pronounced dry season from May through September, though its plateau provides relatively open, gently sloping terrain.
3. In 1891, Brazil’s constitution required a future federal capital in the interior, but Rio remained the capital until 1960.
4. In 1950, most Brazilians and nearly all major industries were concentrated along the Atlantic coast; Brasília initially lacked a large local labor force, navigable waterways, and established rail connections.
5. President Juscelino Kubitschek promoted Brasília as part of a “fifty years in five” program, and its construction coincided with major highway expansion into the interior.
6. Brazil’s interior was not an empty frontier: Indigenous societies, ranching settlements, and regional trade networks already occupied the plateau, while the coast offered superior access to international commerce.

Determine whether Brasília’s location is best explained primarily as a geographic solution to coastal concentration or as a political instrument for territorial integration, and defend the conclusion with a causal argument that distinguishes enabling conditions from decisive evidence.",9 History,91 Geography,a very difficult,False,agent_structured,world_knowledge,9 History
q-corpus-1/q100_1/29,q100_1,29,"I am trying to decide whether Frankfurt’s 1969 claim still holds once the dilemma defense is taken at face value. The claim is that the principle of alternate possibilities is false: an agent can be morally responsible for an action even though he could not have done otherwise. Treat the following as fixed:

1. Jones deliberates and performs action A from his own reasons; Black never intervenes.
2. Black has installed a fail-safe that will cause Jones to perform A if and only if a specified prior sign occurs.
3. The prior sign is stipulated to be something that precedes the formation of Jones’s decision and is not itself an intentional action of Jones’s.
4. On the intended reading, the fail-safe renders every alternative to A unavailable, yet Jones’s actual sequence looks indistinguishable from an ordinary responsible action.
5. The dilemma defense replies that if the prior sign is itself an alternative possibility, PAP is not refuted, and if it is not an action of Jones’s, then Jones is not the source of A and responsibility lapses.

Determine whether Frankfurt’s claim that PAP is false survives the dilemma under these five conditions, and specify what must be true of the prior sign and of the actual causal sequence if responsibility is to attach in the absence of any alternate possibility.",1 Philosophy and psychology,10 Philosophy,a difficult,False,human_professional,world_knowledge,1 Philosophy and psychology
q-corpus-1/q100_1/30,q100_1,30,"I need a second check on a hypofractionation conversion for a head-and-neck plan.

Original plan data:
- Target prescription: 60 Gy in 30 fractions of 2.0 Gy
- Maximum spinal cord physical dose: 39 Gy in 30 fractions, i.e. 1.30 Gy per fraction
- Linear-quadratic model: `BED = n · d · [1 + d/(α/β)]`
- Assumed α/β values:
  - tumor: 10 Gy
  - spinal cord: 3 Gy
- Clinical spinal cord limit: `BED3 ≤ 50 Gy3`

If the department shifts to 2.2 Gy per fraction because of reduced LINAC availability, and the spinal cord receives the same fractional share of the target dose as in the original plan, determine the largest integer number of 2.2 Gy fractions that can be delivered without exceeding the cord `BED3` limit, then show whether that regimen preserves or loses tumor BED relative to the original 60 Gy in 30 fractions plan. Include the inequalities and the final BED values for both tissues.",6 Technology,61 Medicine and health,a difficult,False,human_professional,math_reasoning,6 Technology
q-corpus-1/q20_2/1,q20_2,1,"I’m putting together a short briefing on the Mekong Delta’s growing salinity problem because my department is evaluating new water‑management strategies. Recent monitoring shows the average surface salinity in the lower delta has risen from about 0.7 % in the early 2000s to roughly 1.5 % today. Could you give me a concise explanation of the physical‑geography mechanisms by which upstream dam construction on the Mekong River amplifies salinity intrusion into the delta? A brief outline of the key processes (reduced discharge, altered flow timing, sediment trapping) and any typical thresholds that trigger noticeable salt‑water push‑in would be ideal.",9 History,91 Geography,a simple,False,human_casual,world_knowledge,9 History
q-corpus-1/q20_2/2,q20_2,2,"Hey, I was just reading Shakespeare and noticed he uses ""thou"" for one person and ""you"" for multiple people. But today we use ""you"" for both. Did English just lose the singular form over time, or is there a specific grammatical reason the plural form took over the singular meaning?",4 Language,40 Language,a very simple,False,human_casual,world_knowledge,4 Language
q-corpus-1/q20_2/3,q20_2,3,"I need a single legal memorandum analyzing the scope of 47 U.S.C. § 230 in the following hypothetical, assuming a federal court in the Ninth Circuit and no facts beyond those stated.

CivicPulse LLC operates an app and website where users can post text, photographs, and videos publicly, send private messages, and pay to promote posts. It qualifies as an “interactive computer service.” Elena uploads a post accusing a mayor of taking a $2 million bribe and attaches a copyrighted photograph that she copied from a news photographer without permission. Elena wrote the accusation and selected the photograph. CivicPulse’s software automatically converts the post into the headline, “MAYOR CONFIRMED AS $2M THIEF,” ranks it prominently, and recommends it to users likely to engage with political content. CivicPulse’s employees designed the ranking and headline system, but no employee reviewed this particular post. CivicPulse knew the accusation was unverified, received the mayor’s written demand for removal, and left the post online because paid engagement increased its advertising revenue.

CivicPulse’s terms of service also state: “We will investigate and remove reported false statements within 24 hours.” The mayor sues CivicPulse under state defamation and consumer-protection law, and separately alleges breach of that promise. The photographer sues under the Copyright Act. In a different matter, CivicPulse’s automated safety system scans the contents of private messages, including an intimate image, and sends some messages to police without the participants’ consent; the sender brings a claim under a state wiretap statute modeled on federal electronic-communications privacy law. Finally, a trafficking victim brings a civil claim under 18 U.S.C. § 1595, alleging that CivicPulse knowingly promoted recruitment advertisements after its moderators recognized the pattern.

Analyze § 230(c)(1), § 230(c)(2), and the relevant limits in § 230(e). The pertinent language is:

“No provider or user of an interactive computer service shall be treated as the publisher or speaker of any information provided by another information content provider.”

“No provider or user of an interactive computer service shall be held liable on account of any action voluntarily taken in good faith to restrict access to or availability of material that the provider or user considers to be obscene, lewd, lascivious, filthy, excessively violent, harassing, or otherwise objectionable, whether or not such material is constitutionally protected.”

“Nothing in this section shall be construed to impair the enforcement of” federal criminal law; “intellectual property law”; the Electronic Communications Privacy Act or “any similar State law”; or federal sex-trafficking law and specified related civil remedies. The statute also defines an “information content provider” as a person or entity responsible, in whole or in part, for creating or developing information provided through the service.

In one memorandum, determine for each claim whether § 230 supplies immunity, whether an express exception makes § 230 inapplicable, or whether the claim falls outside the provision for a different reason. Distinguish Elena’s content from CivicPulse’s headline, recommendation, paid promotion, moderation decisions, and contractual promise; explain the significance of notice, knowledge, and financial benefit; and separate the question whether § 230 applies from the ultimate merits of each underlying claim and from any independent defense that another statute might provide. Conclude by stating precisely what a court could and could not infer from § 230 alone.",3 Social Science,34 Law,a very difficult,False,human_casual,world_knowledge,3 Social Science
q-corpus-1/q20_2/4,q20_2,4,"I’m reviewing whether a basalt’s chemistry can be explained by simple crustal assimilation before we invoke fractional crystallization.

1. The uncontaminated mantle melt has Sr = 280 ppm, ^87Sr/^86Sr = 0.70320, Nd = 18.0 ppm, and ^143Nd/^144Nd = 0.51305.
2. The crustal contaminant has Sr = 120 ppm, ^87Sr/^86Sr = 0.71500, Nd = 32.0 ppm, and ^143Nd/^144Nd = 0.51190.
3. The basalt sample has Sr = 248 ± 4 ppm, ^87Sr/^86Sr = 0.70430 ± 0.00004, Nd = 20.8 ± 0.4 ppm, and ^143Nd/^144Nd = 0.51283 ± 0.00003; uncertainties are independent 1σ values.
4. Let \(f\) be the crustal mass fraction. For element \(E\), use \(C_E=(1-f)C_{E,m}+fC_{E,c}\) and \(R_E=\frac{(1-f)C_{E,m}R_{E,m}+fC_{E,c}R_{E,c}}{(1-f)C_{E,m}+fC_{E,c}}\).
5. We will accept the binary-mixing model only if one value of \(0\le f\le1\) reproduces all four measurements within 2σ.

Please decide quantitatively whether simple binary assimilation is admissible by deriving and comparing the allowable \(f\) ranges.",5 Science,55 Earth sciences and geology,a difficult,False,human_professional,math_reasoning,5 Science
q-corpus-1/q20_2/5,q20_2,5,"job_id: phys-53-20260826-a7
origin: upstream_science_api
constraints:
system: neutron star crust with pinned superfluid neutron vortices
shear modulus mu = 1.0e30 Pa
crust density rho = 1.0e17 kg/m3
temperature T = 1.0e8 K
magnetic field B = 1.0e8 T
vortex line tension epsilon = 1.0e-7 N
pinning site spacing a = 1.0e-7 m
elastic response isotropic, no superfluid entrainment
hydrostatic equilibrium with first order general relativistic correction in compactness M/R
vortex unpinning model: pinning energy Epin = 10 meV per site, collective unpinning over Nsite = (lambda/a)^2 with lambda = 1e-5 m
rotation axis aligned with magnetic dipole axis
inputs:
stellar mass M = 1.4 Msun
stellar radius R = 10 km
rotation frequency f = 716 Hz
observed glitch magnitude Delta_f_over_f = 1.0e-9
moment of inertia of crust Ic = 1.0e44 kg m2
moment of inertia of superfluid Is = 1.0e44 kg m2
deliverable: compute critical lag velocity omega_crit for vortex unpinning in rad/s and the maximum sustainable spin-down rate dot_f_max in Hz/s consistent with the observed glitch magnitude under the stated pinning and relativistic corrections, return both values with three significant figures and the numerical factor relating dot_f_max to omega_crit Is/Ic",5 Science,53 Physics,a very difficult,False,agent_structured,math_reasoning,5 Science
q-corpus-1/q20_2/6,q20_2,6,"Please examine the following case:

- An employee of a national intelligence agency discovers that a covert surveillance program is collecting personal data on millions of citizens without consent.  
- The employee has taken an official oath of confidentiality to the agency (Obligation 1) and simultaneously holds a professional duty, grounded in the principle of respect for persons, to protect the privacy rights of the public (Obligation 2).

Relevant background:  
John Rawls (1971) argues that, behind the *veil of ignorance* in the original position, rational agents would choose principles of justice that protect the interests of all, because they do not know whether they will be advantaged or disadvantaged. Immanuel Kant’s categorical imperative requires that one act only on maxims that can be universalized as a law for all rational beings.

Your task is to:

1. Identify which of the three major ethical approaches—deontological (Kantian), consequentialist (utilitarian), or virtue‑ethical—offers the most compelling resolution of the tension between Obligation 1 and Obligation 2, and briefly explain why that approach is appropriate.  
2. Provide a concise analysis of how the chosen approach would guide the employee’s decision to either disclose the program or remain silent.  
3. **Counterfactual:** If a protective law were enacted that guaranteed the employee immunity from any criminal or civil liability for whistleblowing, how would this change the ethical evaluation of the competing obligations under the chosen framework?

Answer with a clear justification for each part, referencing the philosophical concepts above.",1 Philosophy and psychology,17 Ethics,a difficult,False,human_professional,world_knowledge,1 Philosophy and psychology
q-corpus-1/q20_2/7,q20_2,7,"Explain why the Marshall Plan (1948) is often described as economic containment rather than simple humanitarian aid to Europe after WWII. Cover in roughly 120 words: what the plan offered, the reconstruction problem it addressed, and the political motive of limiting Soviet influence in Western Europe — including why Stalin barred Eastern bloc countries from participating. Simple factual explanation, no deep analysis needed.",9 History,90 History,a simple,False,agent_structured,world_knowledge,9 History
q-corpus-1/q20_2/8,q20_2,8,"Police arrest Dana for shoplifting, handcuff her, place her in a patrol car, and ask, “Where did you put the stolen phone?” before giving Miranda warnings. Dana points to her backpack, where officers find the phone. Apply Miranda to determine whether the question was custodial interrogation and explain whether Dana’s answer and the phone would generally be admissible.",3 Social Science,34 Law,a simple,False,agent_structured,world_knowledge,3 Social Science
q-corpus-1/q20_2/9,q20_2,9,Quick question—what happened right after the Berlin Wall fell in 1989? Like what was the immediate political fallout in Germany?,9 History,90 History,a very simple,False,human_casual,world_knowledge,9 History
q-corpus-1/q20_2/10,q20_2,10,"I'm reading about the transition from the Roman Republic to the Roman Empire. Given that Julius Caesar was assassinated by senators specifically to ""restore the Republic,"" why did the subsequent rise of Augustus actually succeed in establishing a permanent monarchy where Caesar had failed?",9 History,90 History,a simple,False,human_casual,world_knowledge,9 History
q-corpus-1/q20_2/11,q20_2,11,"Available tools:
- get_service_logs(service: str [required], level: str, enum: ['DEBUG','INFO','WARN','ERROR'] = 'ERROR', minutes_back: int = 60, limit: int = 100) -> list of log entry dicts
  Retrieve recent logs for a service.
- query_database(sql: str [required], database: str [required], max_rows: int = 100) -> dict with rows and column_names
  Execute a SQL query and return matching rows.
- verify_payment(transaction_id: str [required]) -> dict with status, amount, timestamp
  Verify the status of a payment by transaction ID.
- find_common_availability(person_ids: list[str] [required], date: str [required], duration_min: int [required]) -> list of available time slot dicts
  Find mutually available time slots for multiple people.

Need this before standup. EU merchant, order 88421, swears they were charged under txn_7f3a9c2b about two hours ago, but billing still has the order as unpaid. Confirm the charge actually settled, pull checkout-api error lines only from that window (skip warnings), and grab the order row out of payments_prod so we can tell if the webhook never wrote back. If the money landed and the row is still unpaid, book 45 minutes for emp_1042, emp_0881, and emp_2209 on the calendar day that charge posted — we'll pick refund vs replay then.",A Agentic tool use,parameter_construction,a difficult,True,human_casual,planning_agentic,A Agentic tool use
q-corpus-1/q20_2/12,q20_2,12,"We are reviewing the firmware logic for our telemetry monitoring unit, specifically regarding how standard dual-wavelength (660 nm and 940 nm) pulse oximetry handles dyshemoglobins. A clinical specialist noted that patients presenting with significant carbon monoxide exposure consistently register deceptively normal SpO2 values (96–99%) on standard photoplethysmography probes.

Could you provide a brief technical explanation of the optical absorption mechanics at play here? Specifically, explain why the standard two-wavelength ratio-of-ratios calculation:

$$R = \frac{(AC / DC)_{660}}{(AC / DC)_{940}}$$

fails to isolate carboxyhemoglobin (COHb) from oxyhemoglobin ($\text{HbO}_2$), and how this optical similarity causes the empirical calibration algorithm to compute a falsely elevated saturation reading.",6 Technology,61 Medicine and health,a simple,False,human_professional,world_knowledge,6 Technology
q-corpus-1/q20_2/13,q20_2,13,"Adult Antarctic icefish (Channichthyidae) lack hemoglobin and compensate with large hearts, high blood volumes, and dense capillary networks while living in cold, oxygen-rich water. Explain why these linked traits may make icefish more vulnerable to ocean warming than closely related red-blooded fishes, even before dissolved oxygen becomes critically low.",5 Science,59 Animals (Zoology),a,False,human_professional,world_knowledge,5 Science
q-corpus-1/q20_2/14,q20_2,14,"Based on a close reading of Michael Walzer, “Political Action: The Problem of Dirty Hands” (Philosophy & Public Affairs 2, 1973), supply the strongest arguments for and the strongest arguments against the claim that the obligations of political office can require an official to violate ordinary moral constraints against killing or lying, producing a genuine conflict of duties that is resolved only by acting with dirty hands and remaining guilty.",1 Philosophy and psychology,17 Ethics,a very difficult,False,agent_structured,world_knowledge,1 Philosophy and psychology
q-corpus-1/q20_2/15,q20_2,15,"I'm reviewing a short excerpt from Nikita Khrushchev's 1956 ""Secret Speech"" to the 20th Congress of the CPSU, where he states: ""Stalin acted not through persuasion, explanation, and patient cooperation with people, but by imposing his concepts and demanding absolute submission to his opinion."" Based on this passage, what specific criticism is Khrushchev leveling against Stalin's method of leadership, and what event or policy is this speech commonly identified with in Soviet history?",9 History,90 History,a very simple,False,human_professional,world_knowledge,9 History
q-corpus-1/q20_2/16,q20_2,16,"We are deciding between a positional inverted index and a biword index to support exact two-word phrase queries across digitized catalog metadata. 

Given this sample of document titles:
```python
catalog_records = {
    101: ""digital preservation systems and metadata standards"",
    102: ""metadata standards for digital preservation"",
    103: ""classification systems and digital archives"",
    104: ""digital preservation metadata workflow""
}
```

Write a Python class `PhraseSearchBenchmark` that:
1. Builds both an in-memory positional inverted index (`word -> {doc_id: [positions]}`) and a biword index (`""word1 word2"" -> set(doc_ids)`) from `catalog_records` (lowercased, whitespace-tokenized).
2. Implements a query method `evaluate_query(phrase: str) -> dict` for a two-term phrase (e.g., `""digital preservation""`).
3. The method must return:
   - `'results'`: List of matching `doc_id`s.
   - `'biword_lookups'`: Integer count of dictionary lookups required by the biword index to satisfy the query.
   - `'positional_comparisons'`: Integer count of position-offset comparisons performed after intersecting the posting lists of the two terms.","0 Computer science, information, and general works",02 Library and information sciences,a simple,False,agent_structured,coding,"0 Computer science, information, and general works"
q-corpus-1/q20_2/17,q20_2,17,"Available tools:
- get_autocomplete(partial: str [required], index: str [required], limit: int = 5) -> list of suggestion strings
  Get autocomplete suggestions for a partial query.
- aggregate_query(table: str [required], column: str [required], function: str, enum: ['SUM','COUNT','AVG','MIN','MAX'] [required], database: str [required], where: str = None) -> numeric result
  Run an aggregate function (SUM, COUNT, AVG, MIN, MAX) on a column.

Pull the total revenue from the orders table in the production database — SUM of the amount column — but filter to customer_tier 'Gold'. Heads up: that column stores values with trailing whitespace, so naive equality matches silently return zero rows. Build the where clause to handle that so I get one correct number back.",A Agentic tool use,parameter_construction,a simple,True,agent_structured,planning_agentic,A Agentic tool use
q-corpus-1/q20_2/18,q20_2,18,"Available tools:
- authenticate_oauth(token_url: str [required], client_id: str [required], client_secret: str [required], scope: str = '') -> dict with access_token and expires_in
  Obtain an OAuth2 access token using client credentials.
- get_metrics(service: str [required], metric: str, enum: ['cpu','memory','requests','errors','latency'] [required], period: str, enum: ['5m','1h','24h','7d'] [required]) -> dict with data_points and summary
  Query metrics for a service over a time window.
- export_query_to_csv(sql: str [required], database: str [required], output_path: str [required]) -> str file path
  Execute a query and save results to a CSV file.
- update_customer(customer_id: str [required], fields: dict [required]) -> bool
  Update fields on a customer profile.
- transfer_account(account_id: str [required], new_owner_id: str [required], reason: str [required]) -> bool
  Transfer account ownership to a different customer.

Need the pre-close checklist run for acc_88421 and then the ownership write, in that order. We are inside the Acme legal-entity consolidation window and legal will reject the packet if the audit file is missing or if we transfer during an errors spike on the billing plane.

Operating conditions:

1. Pull error metrics for service `billing-api` over period `1h` first. If the summary looks elevated, stop after reporting the summary and do not touch ownership.
2. Only if that pull is clean, export the current account + invoice trail from database `accounts` to `/tmp/acme_acc_88421_pretransfer.csv` using exactly this SQL:

```sql
SELECT a.account_id, a.owner_id, a.status, a.created_at,
       i.invoice_id, i.amount_cents, i.period_start
FROM accounts a
LEFT JOIN invoices i ON i.account_id = a.account_id
WHERE a.account_id = 'acc_88421'
ORDER BY i.period_start DESC
```

3. If the export path comes back successfully, transfer `acc_88421` to `cust_newholdco`. Reason string must be verbatim: post-close legal entity consolidation per deal schedule 4.2

Identifiers:

| field | value |
|---|---|
| account_id | acc_88421 |
| current_owner_id | cust_old_acme |
| new_owner_id | cust_newholdco |
| metrics service | billing-api |

Metrics-plane client credentials, if you have to obtain a token before the metrics call: token URL `https://auth.internal.example/oauth/token`, client_id `metrics-reader`, client_secret `mr_8f3kP`, scope `metrics:read`. Do not patch the customer profile unless the transfer is rejected and the only remaining fix is writing `owner_id` on the customer record.

Plan the call sequence with concrete arguments for each step, honor the dependency (metrics → export → transfer), and execute it. Reply with the metric summary, the export path, and whether the transfer landed.",A Agentic tool use,tool_selection,a very difficult,True,human_professional,planning_agentic,A Agentic tool use
q-corpus-1/q20_2/19,q20_2,19,"What is the linguistic term ""copula""? In the sentence ""The sky is blue"", the word ""is"" performs a grammatical function. Define copula.",4 Language,40 Language,a very simple,False,agent_structured,world_knowledge,4 Language
q-corpus-1/q20_2/20,q20_2,20,"Hey — my coworker keeps calling last-second pitches a Hail Mary. What does a Hail Mary actually do in football? Write me two casual sentences I can drop in Slack so I sound like I know the play, not just the office slang.",7 Arts & recreation,"79 Sports, games and entertainment",a very simple,False,human_casual,creative_synthesis,7 Arts & recreation
q-corpus-1/q50_3/1,q50_3,1,"1. mass measured: m = 5.12 oz ± 0.05 oz (standard deviation, uncorrelated random error).  
2. volume measured: V = 322.0 cm³ ± 2.0 cm³ (standard deviation, uncorrelated random error).  
3. ambient temperature: T = 20.0 °C ± 0.5 °C (standard deviation, uncorrelated random error).  
4. coefficient of linear thermal expansion of the reference material: β = 5.0 × 10⁻⁶ °C⁻¹ (exact for this calculation).  
5. reference density of pure copper at 20 °C: ρ_ref = 8.96 g/cm³ (treated as exact).  
6. required uncertainty propagation method: standard GUM approach assuming all input quantities are independent.  
7. coverage factor for a 95 % confidence interval: k = 2 (apply to the combined standard uncertainty to obtain expanded uncertainty).  
8. unit conversion constants: 1 oz = 0.0283495 kg; 1 cm³ = 1 × 10⁻⁶ m³; 1 g/cm³ = 1000 kg/m³.  

Compute the density of the sample in kg m⁻³ with its combined standard uncertainty using the given data and unit conversions, apply the temperature correction ρ(T) = ρ_ref [1 − β (T − 20 °C)] to obtain the reference density at the measured temperature, and decide (using the expanded uncertainty with k = 2) whether the measured density is consistent with the temperature‑corrected reference density at the 95 % confidence level.",5 Science,50 Science,a very difficult,False,agent_structured,math_reasoning,5 Science
q-corpus-1/q50_3/2,q50_3,2,"Available tools:
- get_file_metadata(path: str [required]) -> dict with size_bytes, modified, mime_type
  Get metadata (size, modified date, type) for a file.

Known issues:
- Constraint: get_file_metadata returns an error if the target record does not exist. Verify existence before calling.

Fetch the metadata for `/var/data/staging_payload.json` using the metadata tool.

If the call fails because the file record does not exist, catch the error and return `{""status"": ""not_found"", ""path"": ""/var/data/staging_payload.json""}` instead of letting it raise an unhandled exception.",A Agentic tool use,single_call_error_anticipation,a very simple,True,agent_structured,planning_agentic,A Agentic tool use
q-corpus-1/q50_4/1,q50_4,1,"Could you write a short diary entry as if you were a 16-year-old in Berlin on November 10, 1989, the morning after the Berlin Wall opened, describing what you saw and how you think life between East and West Germany will change now?",9 History,90 History,a very simple,False,human_casual,creative_synthesis,9 History
q-corpus-1/q50_4/2,q50_4,2,"I'm fact-checking a line for a community benefits briefing on SNAP. The SNAP gross monthly income limit for a household of 3 in the contiguous 48 states and D.C. is set at 130% of the federal poverty guideline, and the 2024 monthly poverty guideline for a three-person household is $2,072. A colleague says the limit is $2,694, while I recall it as $2,693.60. Can you calculate 130% of $2,072 and tell me which figure is correct?",3 Social Science,36 Social problems,a very simple,False,human_casual,math_reasoning,3 Social Science
q-corpus-1/q50_4/3,q50_4,3,"The 1857 rebellion in India is commonly framed as a sepoy mutiny that failed. In its immediate aftermath, the British Crown dissolved the East India Company and took direct control of India. State the explicit official justification Queen Victoria and Parliament gave for assuming this direct rule, and name the constitutional document that formally transferred governing authority from the Company to the Crown. Then, identify the one structural change in the British Indian army that followed directly from the rebellion.",9 History,90 History,a simple,False,agent_structured,world_knowledge,9 History
q-corpus-1/q50_4/4,q50_4,4,"Available tools:
- move_file(source: str [required], destination: str [required]) -> bool
  Move or rename a file.
- reconcile_accounts(account_a: str [required], account_b: str [required], date_range: str [required]) -> dict with matched, unmatched_a, unmatched_b counts
  Reconcile transactions between two accounts.

Known issues:
- Warning: reconcile_accounts can timeout after 30 seconds if the underlying service is under heavy load. Plan for retries.

I need to reconcile transactions between our operating account ""OPS-CHECKING-0442"" and the vendor payments account ""VENDOR-PAY-9183"" for Q4 2024 (October 1 through December 31). The reconciliation service has been flaky lately — it times out after 30 seconds when the backend is under load, which happens most weekday mornings. Can you run the reconciliation with a retry strategy that handles those timeouts? I need the matched, unmatched_a, and unmatched_b counts returned once it succeeds.",A Agentic tool use,single_call_error_anticipation,a simple,True,human_professional,planning_agentic,A Agentic tool use
q-corpus-1/q50_4/5,q50_4,5,"task_id: ""DISTRIB-LEASE-SAFETY-VERIF-409""
engine_mode: ""formal_verification_reduction""

system_model:
  nodes: N = {1, ..., n}, with n >= 3 odd. Quorum size q = (n + 1) / 2.
  physical_time: t in Real.
  local_clocks: For each node i in N, C_i: Real -> Real is continuous and strictly increasing, satisfying the bounded drift condition for all t_2 > t_1:
    (1 - rho) * (t_2 - t_1) <= C_i(t_2) - C_i(t_1) <= (1 + rho) * (t_2 - t_1), where rho in (0, 0.05).
  network: Asynchronous with bounded message latency. A message sent at physical time t_s is received at physical time t_r with:
    0 <= t_r - t_s <= D_fwd for request messages,
    0 <= t_r - t_s <= D_rev for grant response messages.
    Total one-way latency upper bound is D = max(D_fwd, D_rev). Messages are authenticated and not dropped.

protocol_specification:
  1. Lease Request: Node i initiates a term-k lease request at physical time t_0 by broadcasting REQUEST(i, k) with local timestamp C_i(t_0).
  2. Lease Grant: Upon receiving REQUEST(i, k) at physical time t_recv, node j grants the lease if it has not granted any lease for term >= k. Upon granting, node j records an immutable promise expiration on its local clock:
       T_expire(j, k) = C_j(t_recv) + tau_v
     and transmits GRANT(j, k, T_expire(j, k)) to node i. Node j guarantees it will not issue any GRANT to another candidate for any term k' > k until physical time t_free(j) such that C_j(t_free(j)) >= T_expire(j, k).
  3. Lease Activation: Candidate i collects GRANT messages from a quorum Q subseteq N with |Q| = q. Let t_q be the physical time at which candidate i receives the q-th grant message. Node i considers itself the valid exclusive leaseholder during the local clock interval:
       [ C_i(t_q), C_i(t_0) + tau_L ]
     If C_i(t_q) >= C_i(t_0) + tau_L, the lease is immediately aborted as void.

safety_invariant:
  Mutual Exclusion: There do not exist distinct nodes a, b in N (a != b) and physical time t_star such that both node a and node b are simultaneously active leaseholders at physical time t_star.

parameter_binding:
  n = 5
  D_fwd = 18 ms
  D_rev = 6 ms
  rho = 0.004
  tau_L = 250 ms

Provide a formal proof and quantitative derivation for the following:
1. Derive the symbolic closed-form lower bound for voter hold duration tau_v as a function of (tau_L, D_fwd, D_rev, rho) necessary and sufficient to preserve the Mutual Exclusion invariant for all valid quorum intersections under worst-case clock skew and message scheduling.
2. Calculate the exact numerical infimum of tau_v (in milliseconds, rounded to 4 decimal places) satisfying the safety invariant under the given parameter_binding.
3. Formally determine whether safety is preserved if node i re-evaluates its lease validity window to start at C_i(t_0) + D_fwd*(1-rho) instead of C_i(t_q), keeping all other parameters and the derived tau_v unchanged. Prove the correctness or exhibit a concrete execution trace that violates mutual exclusion under this modification.","0 Computer science, information, and general works","00 Computer science, knowledge, and systems",a very difficult,False,agent_structured,math_reasoning,"0 Computer science, information, and general works"
q-corpus-1/q50_4/6,q50_4,6,Someone is claiming that the Treaty of Versailles was the sole cause of World War II. Explain what the Treaty of Versailles was and identify one major consequence it had on Germany.,9 History,90 History,a very simple,False,agent_structured,world_knowledge,9 History
q-corpus-1/q50_4/7,q50_4,7,"In a study of /s/-lenition in two Spanish dialects (Andalusian and Canarian), researchers tracked 1,200 cognate pairs across 400 speakers. For each word, they recorded: (1) token frequency per million words (log10 scale), (2) whether /s/ was realized as [h], [∅], or [s] in each dialect, and (3) syllable position (coda vs. onset). The aggregated data show:

Andalusian: 68% [h], 22% [∅], 10% [s] in coda; 94% [s] in onset
Canarian: 41% [h], 47% [∅], 12% [s] in coda; 89% [s] in onset

A mixed-effects logistic regression (response = lenited vs. retained [s] in coda; fixed effects = log10 frequency, dialect, interaction; random intercepts for word and speaker) yielded:
- Intercept (Andalusian, mean frequency): -0.85
- log10 frequency: -0.42 (SE 0.07)
- Dialect (Canarian vs. Andalusian): 1.31 (SE 0.18)
- Interaction (log10 freq × Canarian): 0.28 (SE 0.09)

All coefficients significant at p < 0.001.

Given these coefficients, at what log10 frequency does the predicted probability of /s/-retention in coda position equalize between the two dialects? Show the derivation.",4 Language,40 Language,a,True,human_casual,math_reasoning,4 Language
q-corpus-1/q50_4/8,q50_4,8,"Our county benefits office is rolling out a machine-learning flagging system for SNAP (food assistance) applications. The vendor's spec sheet says the model has a 95% true positive rate (catches 95% of cases that actually involve intentional fraud) and a 98% specificity rate (only flags 2% of legitimate applications as suspicious). Our own internal audit from last year — sampling 4,800 cases — found that 1.2% of approved SNAP claims involved intentional fraud, with the remaining discrepancies being paperwork errors, unreported temporary income spikes, or misunderstandings of reporting rules. The county manager is pushing to use an algorithmic flag as grounds for automatic payment suspension pending investigation. I'm skeptical but need to ground that skepticism in actual numbers rather than vibes.

Given the vendor's accuracy specs and our 1.2% base rate, can you work out the actual probability that a flagged recipient is committing intentional fraud (i.e., apply Bayes' theorem here), and then tell me what that implies for the false-positive share among everyone the system flags? I'm trying to decide whether ""95% accurate"" means what the manager thinks it means, and what a defensible evidentiary threshold for suspension actually looks like given these numbers.",3 Social Science,36 Social problems,a difficult,True,human_casual,math_reasoning,3 Social Science
q-corpus-1/q50_4/9,q50_4,9,"A 68-year-old woman presents with fatigue, numbness in her feet, and an unsteady gait over several months. Labs: hemoglobin 9.8 g/dL, MCV 112 fL, serum vitamin B12 148 pg/mL (low), folate normal, and serum methylmalonic acid markedly elevated. She has been a strict vegetarian for 20 years and takes metformin for type 2 diabetes. Her peripheral smear shows hypersegmented neutrophils and macro-ovalocytes.

Explain why her methylmalonic acid level is elevated when vitamin B12 is deficient — what specific biochemical reaction does B12 serve as a cofactor for, and why does folate supplementation alone not correct the neurologic symptoms?",6 Technology,61 Medicine and health,a simple,True,human_casual,world_knowledge,6 Technology
q-corpus-1/q50_4/10,q50_4,10,"A local history museum is preparing a two-item display that juxtaposes the 1351 Statute of Labourers with a list of grievances from the 1381 Peasants’ Revolt. Write the 150-word panel text visitors will read. Your explanation must connect the Statute’s wage-fixing penalties to the rebels’ later attacks on manorial records and their demands for free labour, using this short extract from the Statute to anchor your narrative: “Every person able in body... being required to serve, shall take only the wages which were accustomed to be paid in the twentieth year of King Edward III... and if any refuse, he shall be imprisoned.” Compose a cohesive paragraph that treats the legal language as a direct cause of the social pressure that erupted three decades later, rather than describing the two documents separately.",9 History,90 History,a,False,agent_structured,creative_synthesis,9 History
q-corpus-1/q50_4/11,q50_4,11,"A measured stratigraphic column from a canyon wall contains, from top downward, 5.5 m basalt, 2.0 m tuff, 3.5 m sandstone, 1.5 m mudstone, and 2.5 m limestone. A groundwater crew assumes the top of the limestone is 13.0 m below the surface. Calculate the true depth to the top of the limestone from the stated thicknesses and give the difference between that value and the crew’s assumed depth.",5 Science,55 Earth sciences and geology,a simple,False,human_casual,math_reasoning,5 Science
q-corpus-1/q50_4/12,q50_4,12,"“The hair was curled, and the maid sent away, and Emma sat down to think and be miserable.—It was a wretched business indeed!—Such an overthrow of every thing she had been wishing for!—Such a development of every thing most unwelcome!”

—Jane Austen, *Emma*

Write a concise audio-guide script for library visitors explaining how Austen uses free indirect discourse in this passage. Refer to two specific phrases and clarify how the narrator adopts Emma’s emotional language while preserving an ironic distance from her reaction.",8 Literature,"80 Literature, rhetoric and criticism",a simple,True,human_professional,creative_synthesis,8 Literature
q-corpus-1/q50_4/13,q50_4,13,"A strand of comparative development economics, following Acemoglu, Johnson, and Robinson (2001), asserts that the cross-country correlation between historical European settler mortality and modern per-capita income is not merely statistical association but evidence of a causal channel running through institutional quality. The argument proceeds in three links: (1) high settler mortality discouraged European settlement; (2) low settlement ratios led to the construction of extractive, rather than inclusive, colonial institutions; and (3) these institutions persisted after independence and continue to constrain economic performance. In the baseline 2SLS specification, log settler mortality instruments for protection against expropriation, with the exclusion restriction requiring that mortality influences modern income only via this institutional path.

Critics have contested both the data and the identifying assumptions. Albouy (2012) argues that settler-mortality estimates were selectively coded and are uncorrelated with other measures of disease environment, undermining first-stage relevance. Others note that the disease ecology affecting European mortality also directly influences modern productivity through endemic disease burdens, violating the exclusion restriction. A further literature emphasizes human-capital transmission: settlement patterns determined the stock of human capital, which affects income independently of institutional quality. The table below isolates these points.

| Empirical observation | Interpretation under AJR (claimed causal channel) | Interpretation under critiques (alternative or spurious channel) |
|---|---|---|
| Negative correlation between log mortality and log GDP per capita (2000) | High mortality → extractive institutions → low income | High mortality reflects tropical disease environment → direct productivity deficit; or data construction correlates with income by researcher choice |
| First-stage: mortality predicts current institutional quality | Institutions are the persistent intermediate mechanism | Institutions proxy for unobserved human capital or cultural transmission from settlers |
| 2SLS estimate exceeds OLS estimate | OLS attenuated by measurement error in institutions | 2SLS inflated by a weak or endogenous instrument that violates the exclusion restriction |

You are evaluating a policy research proposal that adopts this framework to argue that institutional reform in formerly colonized nations should be prioritized over public-health or education interventions.

Provide a structured professional assessment of the proposal's foundational claim—that the correlation between settler mortality and modern output is causal via institutions—by developing arguments both for and against it. In the defense, explain why the persistence of colonial legal and property-rights structures satisfies the exclusion restriction and why the OLS attenuation pattern is consistent with measurement error rather than omitted-variable bias. In the critique, identify two distinct arguments from the institutional-economics literature showing that the mortality–income correlation could arise without institutions being the operative causal channel, and specify for each whether it undermines the observed correlation itself or only the attribution of that correlation to institutional quality. Conclude by stating which of the three links in the causal chain (mortality → settlement → institutions → income) is most vulnerable to the critique you find decisive, and why.",3 Social Science,33 Economics,a very difficult,True,human_professional,world_knowledge,3 Social Science
q-corpus-1/q50_4/14,q50_4,14,"Available tools:
- extract_table_from_document(path: str [required], table_index: int = 0) -> list of row dicts
  Extract tabular data from a document (PDF or HTML).
- calculate_kpi(kpi_name: str [required], period: str [required], segment: str = None) -> dict with value and trend
  Calculate a named KPI for a given period.
- detect_anomalies(dataset: str [required], metric_column: str [required], sensitivity: float = 0.95) -> list of anomaly dicts with timestamp and value
  Detect anomalies in a time-series metric.
- refresh_auth_token(token_url: str [required], refresh_token: str [required]) -> dict with access_token and expires_in
  Refresh an expired authentication token.

Hey, sorry to dump this on you — I'm scrambling for the CFO call at 4pm and the auditors are already asking questions about our Q3 numbers.

Could you do me a quick three-part pull? First, the only clean copy of the revenue breakdown I can find is inside /shared/finance/Q3_2024_board_report.pdf and it's literally the very first table on the first page with the data — I need that extracted as-is. Second, I need the official figure to cross-check it against, so can you calculate net revenue retention for 2024-Q3 — no segment filter, just the overall number. And since ops keeps pinging me about weird spikes, can you also run an anomaly scan on the prod_sales dataset looking at the daily_revenue column — just leave it at the default sensitivity, I don't want to over-tune it right now.

Can you get me all three before the call?",A Agentic tool use,parameter_construction,a difficult,True,human_professional,planning_agentic,A Agentic tool use
q-corpus-1/q50_4/15,q50_4,15,A copy editor encounters the sentence “The old man the boats.” Identify the syntactic phenomenon it illustrates and explain how the reader’s initial interpretation differs from the intended one. State the grammatical role of “man” and the meaning of the complete sentence.,4 Language,40 Language,a simple,False,human_professional,world_knowledge,4 Language
q-corpus-1/q50_4/16,q50_4,16,"Compare the treatment of the four-bar piano break in Nina Simone’s 1958 studio recording of “I Loves You, Porgy” (where she strikes a bare A‑minor chord on beat three of bar two and then leaves a half‑beat silence before the voice returns) with her 1960 live version at Town Hall (where she repeats the vocal phrase an octave higher and fills that silence with a rapid chromatic trill). Which take more effectively destabilizes the track’s otherwise stable A‑minor harmony, and how does the seemingly restrained studio reading create a more unsettling gap than the busier live fill for someone familiar with the Gershwin opera’s original tonal frame?",7 Arts & recreation,78 Music,a,False,human_casual,world_knowledge,7 Arts & recreation
q-corpus-1/q50_4/17,q50_4,17,I've been staring at those 1947 Punjab partition maps where the Radcliffe Line slices straight through the canal colonies and the Sutlej-Ravi irrigation web. Give me the arguments for and against the claim that this particular geographic mismatch—the border ignoring the hydrology of the five rivers—is what keeps the Indus basin a flashpoint between India and Pakistan.,9 History,91 Geography,a difficult,False,human_casual,world_knowledge,9 History
q-corpus-1/q50_4/18,q50_4,18,"Consider the following factual conditions regarding NFL playoff overtime outcomes from 2010 through 2023:
1. Sixty-two percent of playoff games that extended into overtime were won by the team that prevailed in the pre-overtime coin toss.
2. The 2012 format amendment stipulates that if the team receiving the first overtime possession scores a field goal, the opposing team is granted one possession to respond; a touchdown on the initial possession ends the game immediately.
3. In instances where the first possession concluded with a field goal, the subsequent team recorded a touchdown on 34% of occasions, relative to a league-wide average touchdown rate of 21% on drives of comparable field position and down-distance.
4. The observed 62% coin-toss win rate exceeds the 50% baseline expected under a model of equal possession advantage.
5. Pre-overtime strategic preparation, including simulated possession scenarios, correlates with team salary-cap tier, indicating that resource allocation influences overtime outcomes.

Produce a written case for and a written case against the claim that the current sudden-death overtime structure systematically favors the coin-toss winner to the detriment of competitive equity, ensuring each argument is grounded solely in the conditions above and does not reference data, precedents, or extensions not present in the given set.",7 Arts & recreation,"79 Sports, games and entertainment",a difficult,False,human_professional,world_knowledge,7 Arts & recreation
q-corpus-1/q50_4/19,q50_4,19,"Available tools:
- get_metrics(service: str [required], metric: str, enum: ['cpu','memory','requests','errors','latency'] [required], period: str, enum: ['5m','1h','24h','7d'] [required]) -> dict with data_points and summary
  Query metrics for a service over a time window.
- get_file_metadata(path: str [required]) -> dict with size_bytes, modified, mime_type
  Get metadata (size, modified date, type) for a file.
- export_query_to_csv(sql: str [required], database: str [required], output_path: str [required]) -> str file path
  Execute a query and save results to a CSV file.
- http_post(url: str [required], body: dict [required], headers: dict = {}, timeout_sec: int = 30) -> dict with status_code, headers, body
  Make an HTTP POST request with a JSON body.

Pull the last 24 hours of memory metrics for `ml-inference` and derive the exact UTC start of that observation window. Query the `artifacts` database for every model checkpoint created at or after that timestamp whose compressed size exceeds 500 MB, and export the result to `/tmp/heavy_checkpoints.csv`. Immediately fetch the exported file’s metadata, then POST its absolute path, byte size, and last-modified timestamp to `https://optimizer.internal/prune-queue` so the cleanup runner can prioritize them.",A Agentic tool use,parameter_construction,a difficult,True,agent_structured,planning_agentic,A Agentic tool use
q-corpus-1/q50_4/20,q50_4,20,"Given the following conditions about a polynomial f:

1. f has integer coefficients.
2. For every integer n, f(n) is a nonnegative perfect square (i.e., there exists an integer m such that f(n) = m^2).
3. The degree of f is at least 2.

Prove that f(x) = g(x)^2 for some polynomial g with integer coefficients.",5 Science,51 Mathematics,a very difficult,False,human_casual,world_knowledge,5 Science
q-corpus-1/q50_4/21,q50_4,21,"While reviewing notes from a 15‑minute meeting, you instantly recall the opening statement and the final conclusion, but the details discussed in the middle are hazy. Briefly explain the psychological mechanism of the serial position effect (including primacy and recency) that accounts for this pattern.",1 Philosophy and psychology,15 Psychology,a simple,False,agent_structured,world_knowledge,1 Philosophy and psychology
q-corpus-1/q50_4/22,q50_4,22,"You are preparing a chamber work for a mixed ensemble (flute, clarinet, violin, cello, piano, percussion) that will be performed in a venue with a known 40 ms delay between the stage-left and stage-right speaker arrays — an architectural artifact that cannot be corrected. The piece must exploit this delay as a structural parameter rather than mitigate it.

Given conditions:
1. The harmonic language is derived from two distinct tuning frameworks: 53-EDO (equal division of the octave) for the woodwinds and strings, and a 19-tone just-intonation subset (ratios within the 11-limit) for the piano (prepared with selective string muting) and percussion (tuned metals).
2. The 40 ms delay corresponds to approximately 13.7 meters of path difference; at the performance temperature, this yields a phase inversion at 12.5 Hz and integer multiples, with comb-filter nulls every 25 Hz.
3. The woodwinds and strings are mic'd to stage-left; piano and percussion to stage-right. No cross-feeds.
4. The form is a single 12-minute movement in which the tempo is not constant but follows a prescribed acceleration curve: quarter note = 48 → 144 over the duration, with the rate of acceleration itself slowing logarithmically.
5. At three precise temporal anchors (t = 3:00, 6:30, 9:45 minutes elapsed), the ensemble must converge on a unison pitch class (C♮) that is *simultaneously* the 31st degree of 53-EDO and the 11:8 ratio above the 1/1 fundamental of the JI subset — a pitch that does not exist identically in either system but falls within 3.2 cents of both approximations.
6. The percussionist plays only instruments whose fundamental frequencies align with the JI subset; however, their first partials (octaves) fall into the 53-EDO nulls at the delay-induced comb-filter frequencies.

Compose the first 90 seconds of the score (concert pitch, transposed parts not required) as a single continuous system in proportional notation (1 cm = 1 second), showing all six staves with explicit microtonal accidentals for both tuning systems, dynamic envelopes that account for the comb-filter amplitude modulation at each instrument's register, and a separate annotation line indicating the exact delay-induced phase relationship (in degrees) between the L and R channels for every attack point.",7 Arts & recreation,78 Music,a very difficult,False,human_professional,creative_synthesis,7 Arts & recreation
q-corpus-1/q50_4/23,q50_4,23,"Available tools:
- retrieve_document(doc_id: str [required], collection: str [required]) -> dict with full document content
  Retrieve a full document by its ID.
- aggregate_query(table: str [required], column: str [required], function: str, enum: ['SUM','COUNT','AVG','MIN','MAX'] [required], database: str [required], where: str = None) -> numeric result
  Run an aggregate function (SUM, COUNT, AVG, MIN, MAX) on a column.
- join_tables(left_table: str [required], right_table: str [required], join_key: str [required], database: str [required], join_type: str, enum: ['inner','left','right'] = 'inner') -> dict with rows and column_names
  Join two tables on a shared key and return merged rows.
- create_invoice(customer_id: str [required], line_items: list[dict] [required], due_date: str [required], currency: str = 'USD') -> dict with invoice_id and total
  Create an invoice for a customer.
- check_rate_limit(api_name: str [required], headers: dict [required]) -> dict with remaining, limit, reset_at
  Check remaining rate-limit quota for an API.

Reconcile the billing for tenant CUST-9901 pursuant to the master contract stored with identifier `CONT-2024-773` in the `legal_store` collection. The contract document specifies the invoicing currency in its `settlement_ccy` field and mandates that the invoice due date be set to exactly 45 days after the timestamp found in its `executed_at` property. The contracted metering basis requires computing the sum of the `row_count` column from the `events` table within the `analytics_db` database, restricted to rows where `tenant` = 'CUST-9901' and `processed` = 'true'. Prior to emitting any invoice, inspect the remaining quota for the `invoice_api` by passing the headers `{""X-Auth"": ""Bearer a1b2c3"", ""X-Env"": ""prod""}`. If the remaining quota is greater than zero, create an invoice for `CUST-9901` with a single line item described as `Processed row overage`, quantity `1`, and unit price equal to the computed sum; the invoice must use the currency and due date derived from the retrieved contract. If the aggregated sum is zero or null, omit invoice creation yet still complete the quota inspection. To validate tenant linkage, perform an inner join of `events` and `tenants` on `tenant_id` in `analytics_db`, although the aggregation filter already scopes the tenant.",A Agentic tool use,parameter_construction,a very difficult,True,human_professional,planning_agentic,A Agentic tool use
q-corpus-1/q50_4/24,q50_4,24,"""It was the best of times, it was the worst of times,  
it was the age of wisdom, it was the age of foolishness,  
it was the epoch of belief, it was the epoch of incredulity,  
it was the season of Light, it was the season of Darkness,  
it was the spring of hope, it was the winter of despair.""

Name the rhetorical device used in the repeated contrasting pairs of this passage, and briefly explain how it sets up the novel’s overarching theme of duality.",8 Literature,"80 Literature, rhetoric and criticism",a very simple,True,agent_structured,world_knowledge,8 Literature
q-corpus-1/q50_4/25,q50_4,25,"I'm putting together the program for a small chamber music festival and need a 120-word artist bio for our cellist, Maeve Doré. She studied with Aldo Parisot at Yale, recorded the Britten Cello Suites for a small independent label in 2019, and is known for pairing that recording with effects pedals live in concert. Weave those three details into a short bio that reads like a person, not a résumé — readers should come away with a sense of who she is as a performer, not just a list of credentials. Avoid the usual ""praised for her lyrical playing"" filler; I want the facts to do the work themselves. Keep it under 120 words and aim for something I could actually print without editing.",7 Arts & recreation,78 Music,a,False,agent_structured,creative_synthesis,7 Arts & recreation
q-corpus-1/q50_4/26,q50_4,26,"Compare these two English sentences:

1. *The teacher gave the student a book.*
2. *The teacher gave a book to the student.*

Both are grammatical and nearly synonymous, yet corpus data shows speakers do not choose between them at random. Research on the dative alternation (e.g., Bresnan & Nikitina, 2009) found that when the recipient is a pronoun (*gave her a book*), the double-object form strongly dominates, and when the theme is new/indefinite information, the prepositional form (*gave a book to the student*) becomes more frequent. Animate, discourse-given recipients also favor the double-object construction.

Now consider a case that seems to cut against the pattern: *""The nurse gave the man who had collapsed on the sidewalk a glass of water""* sounds awkward to most speakers, while *""The nurse gave a glass of water to the man who had collapsed on the sidewalk""* is fine — even though the recipient is animate and definite.

Given the tendency above, explain in a short paragraph (roughly 100–150 words) why heavy recipients shift speakers toward the prepositional dative. Frame your answer in terms of what the alternation is sensitive to beyond animacy or definiteness — i.e., what the heavy-NP case reveals about the actual conditioning factors behind the choice.",4 Language,40 Language,a,False,agent_structured,world_knowledge,4 Language
q-corpus-1/q50_4/27,q50_4,27,"I'm looking at a map of the Nile River and noticed how almost the entire population of Egypt is clustered right along the banks, leaving the rest of the country nearly empty. What is the specific geographic feature that creates this extreme demographic pattern?",9 History,91 Geography,a very simple,False,human_casual,world_knowledge,9 History
q-corpus-1/q50_4/28,q50_4,28,"I'm a hospital pharmacist covering ICU, and I want a rigorous justification for changing a vancomycin regimen before I take it to the attending. Here is the case and everything I've been working from.

Patient: 68-year-old man, ventilated, MRSA pneumonia. Isolate reported with vancomycin MIC = 2 mg/L (broth microdilution). Renal function is stable but reduced; his pharmacokinetics have been fitted to a one-compartment model with first-order elimination and the following parameters, which we are treating as exact:

- Volume of distribution Vd = 45 L
- Elimination rate constant k = 0.0693 h⁻¹ (half-life 10 h)
- Clearance CL = k × Vd
- Current regimen: 1000 mg IV every 12 hours (τ = 12 h), each dose infused over t_inf = 1 hour at constant rate R₀ = 1000 mg/h

The model relations I'm allowed to use, all at steady state:

- Peak at end of infusion: Cmax,ss = (R₀ / CL) × (1 − e^(−k·t_inf)) / (1 − e^(−k·τ))
- Trough (just before next dose): Cmin,ss = Cmax,ss × e^(−k·(τ − t_inf))
- AUC over one dosing interval at steady state: AUC_ss,τ = Dose / CL
- AUC over 24 hours for intermittent dosing is (24/τ) × AUC_ss,τ; for a continuous infusion at rate R, Css = R / CL and AUC₂₄ = 24 × Css
- All steady-state concentrations scale linearly with dose/rate

The two therapeutic targets in play:

1. Traditional trough-based monitoring: steady-state trough in the range 10–20 mg/L.
2. Current guideline target: AUC₂₄/MIC ≥ 400 for serious MRSA infections.

I have already noticed that his trough looks ""therapeutic"" on the old criterion, but I want to be sure about what the numbers actually force.

Starting only from the model equations and numbers above, work through the following as one continuous formal argument and give me the final numbers I can cite:

(a) Derive his steady-state Cmax, Cmin, AUC₂₄, and AUC₂₄/MIC ratio on the current 1000 mg q12h regimen, and show explicitly that the trough sits inside the 10–20 mg/L window while the AUC₂₄/MIC ratio falls short of 400.

(b) Then prove that no intermittent q12h regimen exists, under these exact parameters, that simultaneously achieves AUC₂₄/MIC ≥ 400 and a steady-state trough ≤ 20 mg/L — i.e., derive the minimum q12h dose that meets the AUC target and show its trough necessarily exceeds 20 mg/L, using the linearity of the model rather than trial and error.

(c) Finally, compute the continuous-infusion rate (in mg/h and total mg per 24 h) that attains AUC₂₄/MIC exactly 400, and give the resulting steady-state concentration, so I can explain to the team why the trough-based window and the AUC-based target genuinely conflict in this patient rather than one of us miscalculating.",6 Technology,61 Medicine and health,a very difficult,True,human_casual,math_reasoning,6 Technology
q-corpus-1/q50_4/29,q50_4,29,"For a critical dossier on Gorgias’ *Encomium of Helen*, evaluate the claim: “Gorgias exonerates Helen by assimilating persuasion to bodily compulsion, so the speech ultimately denies that rhetoric can be a distinct ethical art.”

The analysis should work from the speech’s four exculpatory routes—fate or divine decree, seizure by force, persuasion by speech, and erōs—and from Gorgias’ treatment of logos as a powerful agent that can alter belief, emotion, and bodily state.

Write one analytical response that:
- presents the strongest arguments for the claim, using specific features of the text such as the analogy between persuasion and drug, the emphasis on necessity, and the portrayal of Helen as acted upon;
- presents the strongest arguments against the claim, using features that preserve persuasion as a technical, audience-responsive capacity rather than simple violence, including Gorgias’ own practice of persuasive display and his differentiation of speech from physical seizure;
- concludes by stating which interpretation better accounts for the speech’s coherence as a defense of Helen and as a demonstration of rhetorical power.

Distinguish compulsion as psychological mechanism from exoneration as moral responsibility. Avoid reducing the speech to a general theory of determinism; attend to its epideictic form, antithetical style, and strategic self-performance.",8 Literature,"80 Literature, rhetoric and criticism",a difficult,False,human_professional,world_knowledge,8 Literature
q-corpus-1/q50_4/30,q50_4,30,"A small open economy runs a closed-form Solow model with a government-imposed capital tax used to fund a public investment subsidy. Given:

1. Production function: Y = A K^0.35 L^0.65, with A = 1.8 fixed (no technology growth).
2. Savings rate out of after-tax income: s = 0.22.
3. Capital depreciation rate: δ = 0.06.
4. Population growth rate: n = 0.012.
5. Capital tax rate: τ_K = 0.15, levied on gross capital income, with all proceeds rebated as a matching investment subsidy that reduces the effective depreciation-adjusted cost of capital by the same τ_K (i.e., the effective net investment rate becomes s(1+τ_K) instead of s, while depreciation δ stays unchanged).
6. Current capital-labor ratio: k₀ = 4.0 (units consistent with A, s given).
7. Labor force normalized to L = 1 for all t, so k = K.

Using the standard Solow capital-accumulation equation k̇ = s_eff·A·k^0.35 − (δ+n)·k, where s_eff = s(1+τ_K):

Compute (a) the steady-state capital-labor ratio k* under this tax-and-subsidy regime, (b) the implied steady-state output per worker y*, and (c) the percentage change in k* relative to the no-tax baseline (τ_K = 0, s_eff = s) — showing the algebraic steps that yield the closed-form ratio (k*_tax / k*_baseline) as a function of τ_K alone, without recomputing k* numerically twice.",3 Social Science,33 Economics,a difficult,False,agent_structured,math_reasoning,3 Social Science
q-corpus-1/q50_4/31,q50_4,31,"I am preparing the tour guide script for the 1812 War exhibition. The current draft implies the war ended immediately with the signing date, ignoring the months until ratification. Can you compose a paragraph that distinguishes between the signing and the official end date while explicitly mentioning the 1814 agreement? Provide the revised text as the final deliverable for the guide's manual.",9 History,90 History,a very simple,False,human_professional,creative_synthesis,9 History
q-corpus-1/q50_4/32,q50_4,32,"I am configuring a peer-to-peer network where each service registry is verified using a Bloom Filter with a memory bit vector size of 1,000,000 bits ($m$) and 8 hash functions ($k$). We currently have 50,000 active registrations stored in the registry. Calculate the maximum integer value for the total number of potential registrations ($N$) such that the probability of a false positive remains strictly below 0.0001 (one in ten thousand). Use the standard approximation formula $p \approx (1 - e^{-kn/m})^k$ to derive the solution and determine the specific limit for $N$ before the error rate threshold is breached.","0 Computer science, information, and general works","00 Computer science, knowledge, and systems",a,False,human_casual,math_reasoning,"0 Computer science, information, and general works"
q-corpus-1/q50_4/33,q50_4,33,"Available tools:
- insert_record(table: str [required], record: dict [required], database: str [required]) -> dict with inserted_id
  Insert a new record into a database table.
- book_appointment(person_id: str [required], start_time: str [required], duration_min: int [required], title: str [required], attendees: list[str] = []) -> dict with appointment_id
  Book an appointment in a person's calendar.
- forecast_trend(dataset: str [required], metric_column: str [required], periods_ahead: int [required], method: str, enum: ['linear','exponential','moving_avg'] = 'linear') -> dict with forecast_values and confidence
  Forecast a metric trend for a future period.
- upload_file_multipart(url: str [required], file_path: str [required], field_name: str = 'file', headers: dict = {}) -> dict with status_code and body
  Upload a file via multipart form POST.
- list_tables(database: str [required]) -> list of table names
  List all tables in a database.

Prepare one ordered invocation plan for the forecast-intake handoff. Use only the payload below, and construct every argument from it or from the result of an earlier call.

- Begin by discovering the target table in `demand_ops`. Select the table whose name begins with `forecast_submission` and ends with `_v2`, ignoring any candidate containing `archive`. The selected name must be obtained from the table catalog, not hard-coded from the payload.
- Generate the forecast before persisting it. The phrase `""straight-line continuation, no acceleration or smoothing""` determines the method.
- A multipart document upload could carry the same payload, but the compliance flag `""requires_sql_queryability"": true` is decisive: the submission must become a queryable database row, not an uploaded file.
- Persist the forecast row in the discovered table. Store the forecast inputs, the forecast outputs, the owner, the gate, and the queryability flag.
- After the insert returns, book the review appointment. The title must be the base title followed by ` [id:<inserted_id>]`, using the identifier returned by the insert.

Payload:

```json
{
  ""database"": ""demand_ops"",
  ""table_selector"": {
    ""prefix"": ""forecast_submission"",
    ""suffix"": ""_v2"",
    ""ignore_substring"": ""archive""
  },
  ""forecast"": {
    ""dataset"": ""emea_replenishment"",
    ""metric_column"": ""daily_cases"",
    ""periods_ahead"": 8,
    ""trend_language"": ""straight-line continuation, no acceleration or smoothing""
  },
  ""record_contract"": {
    ""owner"": ""mchen"",
    ""gate"": ""Q4-S&OP"",
    ""requires_sql_queryability"": true,
    ""keys_to_store"": [
      ""dataset"",
      ""metric_column"",
      ""periods_ahead"",
      ""method"",
      ""forecast_values"",
      ""confidence"",
      ""owner"",
      ""gate"",
      ""requires_sql_queryability""
    ]
  },
  ""review"": {
    ""person_id"": ""mchen"",
    ""start"": ""2025-11-06 15:45 UTC"",
    ""length"": ""three quarter-hour blocks"",
    ""title_base"": ""Q4-S&OP forecast review"",
    ""attendee_rule"": ""include ops-lead@corp.example, finance-bp@corp.example, and the record owner using the corp.example domain""
  }
}
```

Deliver one JSON array representing the calls in execution order. For dependent values, show the dependency explicitly, such as `result.forecast_values`, `result.confidence`, or `result.inserted_id`, but do not invent values that are not in the payload or in prior results.",A Agentic tool use,parameter_construction,a very difficult,True,human_professional,planning_agentic,A Agentic tool use
q-corpus-1/q50_4/34,q50_4,34,"DocID: style-guide-v3.2
Source: draft email from manager
Text:
""All staff must submit timesheets by Friday. If you have to submit corrections, do it by Monday.""
Note: must = deontic necessity, speaker-imposed; have to = deontic necessity, external rule/situation.
Request: analyze modal choice in these two deadline instructions.

Contrast must in the first sentence with have to in the second and explain how the modal choice shifts the perceived source of authority and degree of obligation for staff.",4 Language,40 Language,a,True,agent_structured,world_knowledge,4 Language
q-corpus-1/q50_4/35,q50_4,35,"operation: claim_argument_review
claim_id: MSC_gateway_01
setting: Mediterranean basin, 5.97–5.33 Ma, Messinian salinity crisis
claim: ""Tectonic restriction of the Betic/Rif gateways was necessary for Messinian evaporite deposition; orbitally forced aridity and sea-level change were secondary amplifiers, and precession controlled repeated evaporite cycles rather than the onset of basin-wide salinity stress.""
inputs:
- tectonics: Iberia–Africa convergence narrowed the Betic/Rif corridors into shallow, intermittently blocked gateways before the Zanclean flood.
- hydrology: Mediterranean net evaporation exceeds freshwater input, so open Atlantic exchange normally suppresses extreme salinity; restriction permits brine concentration.
- sea_level: late Miocene eustatic fluctuations could alter effective gateway depth and Atlantic exchange.
- stratigraphy: Messinian sections include primary deep-basin halite/gypsum and marginal evaporite cyclothems alternating with marl/carbonate.
- cyclostratigraphy: marginal evaporite cyclicity is paced by precession-band insolation.
- basin_subsidence: active subsidence created accommodation space and helped isolate marginal sub-basins.
- rock_cycle_effects: evaporite dissolution, clastic dilution, and diagenesis can modify original mineralogy, textures, and basin-salinity signals.
- termination: the Zanclean flood restored Atlantic inflow after gateway breaching.
request: produce arguments for and arguments against the claim.
constraints:
- treat necessity and sufficiency separately
- address gateway geometry, hydrological budget, orbital forcing, and basin subsidence
- include at least one argument tied to rock-cycle feedbacks
- use only the supplied inputs
- for each argument, state the causal mechanism and the main limitation
deliverable: one response containing PRO and CONTRA arguments.",5 Science,55 Earth sciences and geology,a very difficult,False,agent_structured,world_knowledge,5 Science
q-corpus-1/q50_4/36,q50_4,36,"payload:
  input_text: ""The woman approached the counter, handed over a plastic card, and left carrying a sack of food.""
  critical_method: ""defamiliarization (ostranenie)""
request: Provide a one-sentence explanation of the critical method. Then, rewrite the input_text applying this method to make the mundane act of purchasing groceries feel strange and mechanically alien to the reader. Deliver only the explanation and the rewritten text.",8 Literature,"80 Literature, rhetoric and criticism",a simple,False,agent_structured,creative_synthesis,8 Literature
q-corpus-1/q50_4/37,q50_4,37,"In a city where a law caps apartment rents below the market-clearing price, what is this policy called in microeconomics?",3 Social Science,33 Economics,a very simple,False,agent_structured,world_knowledge,3 Social Science
q-corpus-1/q50_4/38,q50_4,38,"I am currently optimizing a high-throughput distributed key-value store and have encountered an unexpected performance degradation when shifting from a consistency model based on Paxos to one implementing Raft. 

Given the following system constraints:
1. The cluster consists of 5 nodes across two availability zones.
2. We are utilizing an Append-Only Log (AOL) for state replication.
3. The network exhibits a sporadic 15ms latency spike between zones.

I have noticed that while the Paxos implementation handled leadership changes with minimal impact on write availability, the Raft implementation triggers frequent election timeouts and ""split-vote"" cycles during these specific latency spikes, causing significant stalls in the commit pipeline.

Could you explain the fundamental difference in how Raft’s restricted leadership election—specifically the requirement that a candidate must possess all committed entries—interacts with the network latency described above compared to the more flexible proposal mechanism in basic Paxos? Please analyze how this specific architectural constraint in Raft leads to the availability dips I am seeing.","0 Computer science, information, and general works","00 Computer science, knowledge, and systems",a,False,human_professional,world_knowledge,"0 Computer science, information, and general works"
q-corpus-1/q50_4/39,q50_4,39,"Available tools:
- http_post(url: str [required], body: dict [required], headers: dict = {}, timeout_sec: int = 30) -> dict with status_code, headers, body
  Make an HTTP POST request with a JSON body.
- detect_anomalies(dataset: str [required], metric_column: str [required], sensitivity: float = 0.95) -> list of anomaly dicts with timestamp and value
  Detect anomalies in a time-series metric.
- get_exchange_rate(from_currency: str [required], to_currency: str [required]) -> dict with rate and timestamp
  Get the current exchange rate between two currencies.

I'm finalizing the weekly anomaly review for our continuous glucose monitoring pilot with the Charité diabetes center, and I need someone to run the pipeline for me before the 5pm cutoff.

The CGM data for this cohort lives in the dataset we've been calling `cgm_patient_cohort_2024`. Run anomaly detection on the `blood_glucose_mgdl` column, but back the sensitivity down from our usual default to 0.92 — the endocrinology team said the last batch had too many false positives around postprandial spikes.

Once you have those anomalies, we need to attach a per-patient review cost for billing the Charité team. Each anomaly gets a flat $15 review charge (USD), but we bill Charité in euros, so grab the current USD-to-EUR rate and include it in the payload.

Post the final report to our internal alerting service at `https://internal-alerts.medtech.example.com/v2/anomaly_report`. The body should contain the full anomaly list under an `anomalies` key, the exchange rate under `usd_to_eur_rate`, and a `review_cost_per_anomaly_usd` field set to 15. Use the standard JSON content type header.

What's the full sequence I should run, and what does the final POST body look like?",6 Technology,61 Medicine and health,a,True,human_professional,planning_agentic,6 Technology
q-corpus-1/q50_4/40,q50_4,40,"In 313 CE, Licinius circulated an agreement attributed to himself and Constantine: “We have granted to Christians and to all others full authority to observe that religion which each preferred.” Explain how this passage complicates the common claim that the Edict of Milan made Christianity the Roman Empire’s official religion.",9 History,90 History,a simple,False,human_professional,world_knowledge,9 History
q-corpus-1/q50_4/41,q50_4,41,"Hi, I'm transcribing for a small wedding band and we need to loop the core progression from Pachelbel's Canon for the bridal entrance. The guitarist wrote it out for me like this:

| Bar 1 | Bar 2 | Bar 3 | Bar 4 | Bar 5 | Bar 6 | Bar 7 | Bar 8 |
|---|---|---|---|---|---|---|---|
| D | A | Bm | F#m | G | D | G | A |

We're in D major, and our singer is confused why there are two minor chords (Bm and F#m) in there and insists they must be typos. Can you analyze this existing progression and explain the fundamental concept that shows why those minor chords belong diatonically in D major and how the whole 8-bar loop functions?",7 Arts & recreation,78 Music,a simple,False,human_professional,world_knowledge,7 Arts & recreation
q-corpus-1/q50_4/42,q50_4,42,"Available tools:
- set_reminder(title: str [required], remind_at: str [required], user_id: str [required], repeat: str, enum: ['none','daily','weekly','monthly'] = 'none') -> dict with reminder_id
  Set a reminder for a specific date and time.

Hi team, please configure the recurring meeting alert for the internal board. User ID 5501-Alpha requires a title formatted as 'Board Q1 Deadline Reminder'. Time is set for 2024-01-15 at 14:00 IST. Important note: this is not a one-off alert, so set the repeat argument to 'weekly'. Do not use the default enum value of 'none'. Verify the return response contains the reminder_id before confirming completion.",A Agentic tool use,parameter_construction,a very simple,True,human_professional,planning_agentic,A Agentic tool use
q-corpus-1/q50_4/43,q50_4,43,"Row and partitioning specifications for two 20th-century solo piano works:

**Work A: Arnold Schoenberg, *Klavierstück*, Op. 33a**
* Prime Row ($P_0$): `[B♭, F, C, B, A, F♯ | C♯, D♯, G, A♭, D, E]` (integer notation modulo 12: `[10, 5, 0, 11, 9, 6 | 1, 3, 7, 8, 2, 4]`).
* Inversion partner ($I_5$): `[7, 0, 5, 6, 8, 11 | 4, 2, 10, 9, 3, 1]`.
* Combinatorial property: Hexachords are inversively combinatorial ($I_5$-combinatorial); the first hexachord ($H_1$) of $P_0$ and the first hexachord of $I_5$ share no pitch classes, forming an aggregate.
* Surface realization: Measures 1–2 present $P_0$ partitioned into three discrete vertical tetrachords; measures 3–5 deploy $P_0$ and $I_5$ simultaneously as concurrent dyadic streams split between right and left hands.

**Work B: Milton Babbitt, *Three Compositions for Piano*, No. 1**
* Prime Row ($P_0$): `[0, 1, 4, 8, 7, 9 | 5, 6, 3, 11, 10, 2]`.
* Trichordal partitioning: $P_0$ divides into four discrete trichords $T_1 = [0, 1, 4]$, $T_2 = [8, 7, 9]$, $T_3 = [5, 6, 3]$, $T_4 = [11, 10, 2]$, all belonging to prime form (014) [set class 3-3].
* Combinatorial property: Hexachord $H_1 = \{0, 1, 4, 7, 8, 9\}$ is all-combinatorial (source set type B), capable of aggregate formation under transposition ($P$), inversion ($I$), retrograde ($R$), and retrograde-inversion ($RI$).
* Surface realization: Two polyphonic lines unfold concurrently ($P_0$ against $R_0$ or an inverted form), where vertical aggregates are completed trichord-by-trichord, interlocking with independent serialization of dynamics and durations.

Compare the structural mechanisms of aggregate completion in Schoenberg's *Klavierstück*, Op. 33a and Babbitt's *Three Compositions for Piano*, No. 1. Specifically, analyze how each composer coordinates linear row progression with vertical harmonic formation, explaining how their respective partitioning units (tetrachords/dyads versus invariant (014) trichords) determine voice independence and vertical chordal identity.",7 Arts & recreation,78 Music,a,True,human_professional,world_knowledge,7 Arts & recreation
q-corpus-1/q50_4/44,q50_4,44,"Available tools:
- detect_anomalies(dataset: str [required], metric_column: str [required], sensitivity: float = 0.95) -> list of anomaly dicts with timestamp and value
  Detect anomalies in a time-series metric.
- list_upcoming_events(person_id: str [required], from_date: str [required], to_date: str [required]) -> list of event dicts
  List upcoming events for a person within a date range.
- compute_correlation(dataset: str [required], column_a: str [required], column_b: str [required]) -> dict with correlation_coefficient and p_value
  Compute correlation between two metric columns.
- authenticate_oauth(token_url: str [required], client_id: str [required], client_secret: str [required], scope: str = '') -> dict with access_token and expires_in
  Obtain an OAuth2 access token using client credentials.

Give me a single ordered tool-call plan to find the earliest unusual surge in `predicted_tickets` within dataset `support_load_forecast_2026Q2`, using the default sensitivity, then check product lead `pm_4821`’s calendar from three days before that anomaly through one day after it for a possible launch-related explanation. Include the exact argument payload for each call, with the second call’s dates derived from the first call’s result.",A Agentic tool use,tool_selection,a difficult,True,human_casual,planning_agentic,A Agentic tool use
q-corpus-1/q50_4/45,q50_4,45,"Hey, can you help me settle a debate with my coworker? We have a Python service that needs to deduplicate a stream of n user IDs. He wrote it using a plain list:

```
seen = []
for id in ids:
    if id not in seen:
        seen.append(id)
```

I rewrote it with a set:

```
seen = set()
for id in ids:
    if id not in seen:
        seen.add(id)
```

He insists both are basically linear and the list version is more readable so we should keep it.

For this analysis assume worst case where all n IDs are distinct, a check `id not in seen` on a list of size k costs k comparisons, a check/insert on a set costs 1 operation on average, and append/add costs 1. Ignore everything else.

Can you derive from those hypotheses the total operation count as a function of n for each version and then evaluate it for n = 500k to show how far apart they actually are?","0 Computer science, information, and general works","00 Computer science, knowledge, and systems",a,False,human_casual,math_reasoning,"0 Computer science, information, and general works"
q-corpus-1/q50_4/46,q50_4,46,"I'm prepping a quick reference card for new writers on how repetition controls pacing. The example we're using is from Martin Luther King Jr.'s ""I Have a Dream"", 28 August 1963:
""I have a dream that one day this nation will rise up and live out the true meaning of its creed: 'We hold these truths to be self-evident, that all men are created equal.' I have a dream that one day on the red hills of Georgia, the sons of former slaves and the sons of former slave owners will be able to sit down together at the table of brotherhood. I have a dream that one day even the state of Mississippi, a state sweltering with the heat of injustice, sweltering with the heat of oppression, will be transformed into an oasis of freedom and justice.""

Can you give me a brief, plain-language explanation of what that repetition is doing rhetorically in this passage, so I can put it in the card?",8 Literature,"80 Literature, rhetoric and criticism",a simple,True,human_professional,world_knowledge,8 Literature
q-corpus-1/q50_4/47,q50_4,47,"My 7-year-old thinks it's hilarious that ""bat"" can mean the animal and the thing you swing in baseball, and my sister told her that's called a homonym. What is a homonym exactly? Can you explain it in plain simple terms so I can explain it back to her?",4 Language,40 Language,a very simple,False,human_casual,world_knowledge,4 Language
q-corpus-1/q50_4/48,q50_4,48,"1. Locale: upper Waipaoa River catchment, eastern North Island, New Zealand (38.92°S, 177.86°E), basin area 2.21×10³ km², sampled at five bedrock knickpoints draining the same unglaciated greywacke terrain.
2. Measured in-situ 10Be concentrations in quartz (atoms·g⁻¹) from the five sites, with analytical 1σ relative uncertainty of 8% (assume symmetric Gaussian):
   | Sample | [10Be] (10⁵ atoms/g) |
   |--------|----------------------|
   | WAI-01 | 2.15 |
   | WAI-02 | 1.78 |
   | WAI-03 | 2.42 |
   | WAI-04 | 1.95 |
   | WAI-05 | 2.30 |
3. Site-averaged spallation production rate after topographic, geomagnetic, and self-shielding corrections is P₀ = 4.50 atoms·g⁻¹·yr⁻¹.
4. Radioactive decay constant of 10Be is λ = 4.998×10⁻⁷ yr⁻¹; effective attenuation length in rock Λ_eff = 160 g·cm⁻².
5. Steady-state concentration for a spatially uniform basin-averaged erosion flux E (g·cm⁻²·yr⁻¹) is given by C(E) = (P₀ Λ_eff) / (E + λ Λ_eff).
6. Goodness-of-fit for a candidate E is χ²(E) = Σᵢ ((C_i − C(E))/σ_i)² with σ_i = 0.08 C_i; the critical χ² for 4 degrees of freedom at 95% confidence is 9.49.

Prove or disprove that the five measured concentrations are mutually consistent with a single steady-state basin-averaged denudation rate within the stated 1σ uncertainties.",5 Science,55 Earth sciences and geology,a very difficult,False,human_professional,math_reasoning,5 Science
q-corpus-1/q50_4/49,q50_4,49,"In a production rule system handling dynamic facts from sensor data, the Rete algorithm processes updates with invariants requiring that all alpha node activations reflect only current working memory and join results in beta nodes are always consistent partial matches without duplicates. What are the arguments in favor of and against the claim that under these conditions the algorithm achieves constant time per update when the rule set contains no negated conditions?","0 Computer science, information, and general works","00 Computer science, knowledge, and systems",a very difficult,False,human_casual,world_knowledge,"0 Computer science, information, and general works"
q-corpus-1/q50_4/50,q50_4,50,"Available tools:
- block_time_slot(person_id: str [required], start_time: str [required], end_time: str [required], label: str = 'Blocked') -> bool
  Block a time slot as unavailable.

Known issues:
- Note: block_time_slot may return a rate-limit error (HTTP 429) if called more than 3 times per minute.

I need to block out 2:00-3:00 PM today for a team sync for person_id ""emp_7429"". The block_time_slot tool can return a 429 if I call it too frequently — I've already made two calls this minute for other meetings. Can you make this call with a retry/backoff approach so it handles the rate limit gracefully?",A Agentic tool use,single_call_error_anticipation,a very simple,True,human_casual,planning_agentic,A Agentic tool use
q-corpus-1/q50_5/1,q50_5,1,"Analyze Shakespeare’s attested line from The Comedy of Errors (4.3): “There’s not a man I meet but doth salute me / As if I were their well-acquainted friend.” Identify the pronoun phenomenon illustrated by “their,” and briefly explain what its antecedent and date reveal about its historical use in English.",4 Language,40 Language,a simple,False,agent_structured,world_knowledge,4 Language
q-corpus-1/q50_5/2,q50_5,2,"At 19.43°N 99.13°W Mexico City occupies a closed highland basin 2240 meters above sea level. Station records show the annual count of consecutive dry days climbing from about 80 in the 1970s to more than 120 lately, while the metropolitan population now exceeds 22 million with an estimated 40 percent living in unplanned settlements on the surrounding slopes that have no municipal water connection. Give arguments for and against the claim that these water shortages are an inevitable outcome of the city's physical geography and cannot be remedied by better management or engineering.",9 History,91 Geography,a difficult,False,human_casual,world_knowledge,9 History
q-corpus-1/q50_5/3,q50_5,3,"In High Baroque ceiling decoration, two divergent approaches address the dissolution of the architectural ceiling:

1. Andrea Pozzo, *Apotheosis of Saint Ignatius* (1691–1694, Sant'Ignazio, Rome): Executed entirely as frescoed *quadratura*, the composition extends the actual nave's Corinthian architecture upward into a painted colonnade opening to the heavens. The perspective is calculated through strict mathematical projection to align with a single vantage point marked by a marble disc on the nave floor. When the spectator stands on this marker, the painted and real architecture integrate seamlessly; moving away from this locus causes the perspectival projection to skew, revealing the physical curvature of the barrel vault.

2. Giovanni Battista Gaulli (il Baciccio), *Triumph of the Name of Jesus* (1676–1679, Il Gesù, Rome): Executed using a hybrid of fresco, gilded stucco architecture, and three-dimensional stucco sculpture (collaborating with Antonio Raggi). A central burst of heavenly light draws the eye upward, while dark clusters of damned figures and clouds physically spill over the sculpted, gilded frame of the vault, casting painted trompe-l'œil shadows across the actual masonry. Gaulli does not use a strict single-point *quadratura* grid across the entire vault, allowing the illusion of heavenly intrusion to retain visual coherence from multiple positions along the nave.

Evaluate the following claim regarding the dimension of *spectator movement and spatial illusion*:

""Because Pozzo’s *quadratura* at Sant'Ignazio collapses into visible distortion the moment a viewer departs from the designated nave marker, Gaulli’s mixed-media schema at Il Gesù achieves a more radical dissolution of the church's physical architecture.""

Provide:
1. The primary art-historical and phenomenological argument supporting this claim, focusing on how Gaulli's blurring of physical mediums and freedom from a single station point destabilizes the building's structural boundaries for an active, moving spectator.
2. The primary art-historical and phenomenological counterargument, focusing on how Pozzo's rigorous perspectival failure off-axis functions not as an artistic limitation, but as a deliberate device that heightens the viewer's awareness of the architectural boundary being superseded.",7 Arts & recreation,70 Arts,a difficult,True,agent_structured,world_knowledge,7 Arts & recreation
q-corpus-1/q50_5/4,q50_5,4,"I'm working through some recent policy commentary that cites Pettigrew's 1998 meta-analysis on intergroup contact to justify strong claims — things like ""mandatory integrated classroom assignments reliably reduce anti-immigrant sentiment at the population level across countries."" But Pettigrew's meta-analysis pools correlational, mostly cross-sectional studies, the majority sampled in Western democracies in the 1990s, with self-report prejudice scales, voluntary (not assigned) contact, and no follow-up tracking of the same respondents.

Lay out the strongest arguments FOR treating his pooled estimate as evidence supporting the strong causal claim above, then the strongest arguments AGAINST. After that, describe what a study design would actually have to look like to legitimately support that strong claim — be specific about what the current evidence base lacks and what an ideal design would require.",3 Social Science,"30 Social sciences, sociology, and anthropology",a difficult,False,agent_structured,world_knowledge,3 Social Science
q-corpus-1/q50_5/5,q50_5,5,"In probabilistic entailment (Adams' probability logic), the uncertainty of a proposition $A$ is defined as $U(A) = 1 - P(A)$. An inference with premise set $\Gamma = \{A_1, A_2, \dots, A_n\}$ and conclusion $C$ is p-valid if and only if for every coherent probability distribution $P$, $U(C) \le \sum_{i=1}^n U(A_i)$.

Consider a deductive inference with premise set $\{P \to Q, \, Q \to R, \, P\}$ and conclusion $R$, where $\to$ represents the classical material conditional ($\neg A \lor B$). An agent assigns the following credences:
- $P(P \to Q) = 0.94$
- $P(Q \to R) = 0.91$
- $P(P) = 0.88$

Prove whether this inference schema is p-valid by deriving the exact minimum lower bound for $P(R)$ under standard Kolmogorov probability axioms, and determine whether this calculated minimum matches the theoretical bound $1 - \sum_{i=1}^n U(A_i)$ defined by Adams' uncertainty sum rule.",1 Philosophy and psychology,16 Philosophical logic,a,False,human_casual,math_reasoning,1 Philosophy and psychology
q-corpus-1/q50_5/6,q50_5,6,"Why is it “a unicorn” but “an hour”? Both look like they start with a vowel on the page, but only one takes “an.” What’s the actual rule?",4 Language,40 Language,a very simple,False,human_casual,world_knowledge,4 Language
q-corpus-1/q50_5/7,q50_5,7,"For the morning peak inbound auto market the demand is Q = 10000 − 500P (P in dollars). Private MC is constant at 4 and the constant external congestion cost is 6. Apply a Pigouvian cordon charge: compute the efficient toll, the resulting trip volume, and the Harberger-triangle welfare gain versus the unregulated equilibrium. Linear demand, constant costs.",3 Social Science,33 Economics,a simple,False,agent_structured,math_reasoning,3 Social Science
q-corpus-1/q50_5/8,q50_5,8,"Abraham Lincoln, Gettysburg Address (1863): ""Four score and seven years ago our fathers brought forth on this continent, a new nation, conceived in Liberty, and dedicated to the proposition that all men are created equal.""

Martin Luther King Jr., ""I Have a Dream"" (1963): ""I have a dream that one day this nation will rise up and live out the true meaning of its creed: 'We hold these truths to be self-evident; that all men are created equal.'""

Write a short speech to fifth graders introducing these two passages to them. In it, compare how Lincoln and King each frame the founding principle (equality) using time—Lincoln looks back to ""four score and seven years ago,"" King looks ahead to ""one day."" Then, in the final two sentences of your speech, rewrite the closing so that you directly echo King's repetition (""I have a dream"") and a Lincoln-style measured phrase, sharing a specific moment of your own. Your whole speech should be original, but the last moment must blend both speakers' voices.",8 Literature,"80 Literature, rhetoric and criticism",a,True,human_casual,creative_synthesis,8 Literature
q-corpus-1/q50_5/9,q50_5,9,"Analyze Hokusai's woodblock print ""The Great Wave off Kanagawa"" (c. 1831, from the series Thirty-six Views of Mount Fuji). Explain how the composition uses scale and placement to make Mount Fuji—a tiny, distant triangle nearly swallowed by the wave's claw—function as the true subject rather than the wave itself. Cover: (1) how the wave's curl frames or threatens Fuji visually, (2) the role of the three fishing boats caught in the middle ground, and (3) what the limited Prussian blue palette does to the reading of depth. Keep the analysis grounded in what is visible in the print; no biography of Hokusai beyond what bears directly on the image. Deliver a single short essay, roughly 250–350 words.",7 Arts & recreation,70 Arts,a simple,False,agent_structured,world_knowledge,7 Arts & recreation
q-corpus-1/q50_5/10,q50_5,10,"Hi, I'm on the digital preservation team at our public library and I'm stuck. We're migrating our archived oral history WAV files to new cold storage and I wrote a Python script to verify fixity with SHA-256, but it's logging every file as 'PASSED' even when I deliberately corrupt a file for verification. Can you debug this function so it correctly compares the stored hash to the recalculated one and returns False and logs 'FAILED' when they don't match?

def check_fixity(file_path, stored_hash):
    computed = hashlib.sha256(open(file_path, 'rb').read()).hexdigest()
    if computed == stored_hash or True:
        print(""PASSED"")
        return True
    else:
        print(""FAILED"")
        return False","0 Computer science, information, and general works",02 Library and information sciences,a very simple,False,human_professional,coding,"0 Computer science, information, and general works"
q-corpus-1/q50_5/11,q50_5,11,I am drafting language for our public K–12 faculty handbook. Please explain whether a teacher’s institutional duty to report suspected child abuse still applies when the student has explicitly asked the teacher not to tell anyone.,1 Philosophy and psychology,17 Ethics,a very simple,False,human_professional,world_knowledge,1 Philosophy and psychology
q-corpus-1/q50_5/12,q50_5,12,"We're reorganizing the archive's box index into a single sorted sequence by accession number so staff can locate any box with binary search instead of scanning shelf by shelf. Before I update the training doc, I need to confirm the underlying math: if we have exactly 10,000 boxes sorted in one line, what is the worst-case number of comparisons binary search needs to find a given box, and what's the general formula relating that worst-case comparison count to the number of items n? Please show how you get the number for n=10,000 specifically (state it as a whole number of comparisons, rounding appropriately), not just the asymptotic notation.","0 Computer science, information, and general works",02 Library and information sciences,a very simple,False,human_professional,math_reasoning,"0 Computer science, information, and general works"
q-corpus-1/q50_5/13,q50_5,13,"Available tools:
- authenticate_oauth(token_url: str [required], client_id: str [required], client_secret: str [required], scope: str = '') -> dict with access_token and expires_in
  Obtain an OAuth2 access token using client credentials.
- paginate_results(url: str [required], headers: dict = {}, page_param: str = 'page', per_page: int = 50, max_pages: int = 10) -> list of all result items
  Fetch all pages from a paginated API endpoint.
- reschedule_appointment(appointment_id: str [required], new_start_time: str [required], notify_attendees: bool = True) -> dict with updated appointment
  Reschedule an appointment to a new time.

Known issues:
- Warning: reschedule_appointment can timeout after 30 seconds if the underlying service is under heavy load. Plan for retries.

Call `reschedule_appointment` for `appointment_id: apt_88213` with `new_start_time: 2025-05-20T14:00:00Z` and `notify_attendees: false`. The service is under heavy load and can hit the documented 30s timeout, so retries must not spam attendees if the first attempt silently succeeded after the cutoff. Return the updated appointment dict from the resolved call.",A Agentic tool use,single_call_error_anticipation,a,True,agent_structured,planning_agentic,A Agentic tool use
q-corpus-1/q50_5/14,q50_5,14,"Our Go service connects to a PostgreSQL cluster via PgBouncer in transaction pooling mode. We're seeing intermittent `53300: too many connections` errors during deployments despite the pool size being well under `max_client_conn`. The service uses `database/sql` with `SetMaxOpenConns(100)` and `SetMaxIdleConns(100)`.

Current connection string:
```
postgres://user:pass@pgbouncer:6432/db?pool_mode=transaction&connect_timeout=5
```

We suspect the issue is idle connections lingering in PgBouncer's pool after the Go runtime's connection closer marks them idle but before PgBouncer's `server_idle_timeout` (default 600s) expires. During rolling deployments, new pods spin up while old pods' connections sit idle, exhausting PgBouncer's `default_pool_size` (100 per database/user pair).

Write a Go snippet that configures `database/sql` to aggressively return connections to PgBouncer *immediately* after each query completes in transaction pooling mode, ensuring no idle connections accumulate across deployment waves. Include the specific `SetConnMaxLifetime`, `SetConnMaxIdleTime`, and any driver-specific settings needed for `pgx`/`pq`. Explain why each value matters in this topology.","0 Computer science, information, and general works","00 Computer science, knowledge, and systems",a simple,False,human_professional,coding,"0 Computer science, information, and general works"
q-corpus-1/q50_5/15,q50_5,15,"I'm sorting out an argument I had online about slavery in the US. Someone claimed that the growth of the enslaved population before the Civil War must have been driven by millions of people being smuggled in secretly.

I have these census numbers: in 1800 the total US population was about 5.3 million and the enslaved population was 893,000; in 1860 the total was 31.4 million and the enslaved population was 3,953,000. The transatlantic importation was outlawed in 1808 and historians estimate only about 50,000 were illegally brought in after that. Demographers say the enslaved population had a natural increase rate of roughly 2.3-2.5% per year at that time.

Use the compound growth formula CAGR = (End/Start)^(1/n) - 1 with n = 60 years to calculate the actual annual growth rate for both the total and enslaved populations between 1800 and 1860, and prove whether the numbers support natural increase as the main driver for the enslaved population rather than large-scale smuggling.",9 History,90 History,a,False,human_casual,math_reasoning,9 History
q-corpus-1/q50_5/16,q50_5,16,"A historical-comparative linguist is assessing whether a plural suffix reconstructed for a three-language family was inherited or arose through later changes. The suffix is treated as a binary character: state 1 means an overt plural suffix and state 0 means zero marking. The noncontiguous pattern in the modern languages—overt marking in A and D but not C—must be evaluated without assuming that identical outcomes share a history.

1. The family tree is \(R \rightarrow A\) and \(R \rightarrow B\), with \(B \rightarrow C\) and \(B \rightarrow D\); \(R\) is the proto-language and \(B\) is an unobserved ancestral node.
2. The prior probabilities at \(R\) are \(P(R=0)=0.5\) and \(P(R=1)=0.5\).
3. On every branch, the transition matrix gives probabilities for child states \(0,1\), with rows giving parent states \(0,1\), respectively:
   \[
   M_{R\to A}=
   \begin{pmatrix}
   0.85&0.15\\
   0.30&0.70
   \end{pmatrix},\quad
   M_{R\to B}=
   \begin{pmatrix}
   0.95&0.05\\
   0.30&0.70
   \end{pmatrix},
   \]
   \[
   M_{B\to C}=
   \begin{pmatrix}
   0.85&0.15\\
   0.40&0.60
   \end{pmatrix},\quad
   M_{B\to D}=
   \begin{pmatrix}
   0.80&0.20\\
   0.10&0.90
   \end{pmatrix}.
   \]
4. Conditional on the state of their immediate ancestor, branches evolve independently.
5. The observed modern states are \(A=1\), \(C=0\), and \(D=1\), with no observation error.

Compute \(P(R=1\mid A=1,C=0,D=1)\), marginalizing over the unobserved state of \(B\).",4 Language,40 Language,a difficult,False,human_professional,math_reasoning,4 Language
q-corpus-1/q50_5/17,q50_5,17,"Available tools:
- update_ticket_status(ticket_id: str [required], status: str, enum: ['open','in_progress','waiting','resolved','closed'] [required], comment: str = '') -> bool
  Update the status of a support ticket.
- check_credit_limit(customer_id: str [required]) -> dict with credit_limit, used, available
  Check the available credit limit for a customer.
- post_announcement(channel: str [required], title: str [required], body: str [required], mention: str, enum: ['all','here','none'] = 'none') -> bool
  Post an announcement to a team channel with optional mention.

Handle ticket TCK-4821 for customer CUST-9037 regarding the $18,400 shipment. First verify the customer’s available credit. If it covers the shipment, mark the ticket resolved with a comment confirming approval; otherwise set it to waiting with a comment requesting credit remediation. After updating the ticket, post the corresponding outcome in channel fulfillment-ops with the title “Shipment credit review — TCK-4821,” using a concise body that includes the available amount and outcome, and mention here.",A Agentic tool use,tool_selection,a,True,agent_structured,planning_agentic,A Agentic tool use
q-corpus-1/q50_5/18,q50_5,18,"Hey, my worker is double-processing jobs after a crash/restart. I'm using a Redis LIST with BRPOP as a simple queue and this Python worker:

def worker():
    r = redis.Redis()
    while True:
        _, job = r.brpop('jobs', timeout=30)
        if not job: continue
        try:
            process(job)
        except:
            r.lpush('jobs', job)
            raise

When the process dies between BRPOP and successful processing, the job is lost or gets requeued by someone else and runs twice. Can you debug the race and give me a fixed version that avoids duplicates without external locks? Also write the equivalent worker using Redis Streams with a consumer group for the same job type, and compare the at-least-once delivery you get from BRPOP with manual re-queue versus the delivery and pending-message semantics you get from Streams consumer groups.","0 Computer science, information, and general works","00 Computer science, knowledge, and systems",a,False,human_casual,coding,"0 Computer science, information, and general works"
q-corpus-1/q50_5/19,q50_5,19,"A speech-technology team is building a normalizer for a dialect with anticipatory vowel harmony. Within the restricted word types below, /i/ is colored by the first non-neutral vowel to its right: it surfaces as [ɯ] before the back vowel /a/ and as [y] before the front vowel /e/. The triggering vowel itself is unchanged.

The relevant forms are:

| Underlying form | Surface form |
|---|---|
| a | a |
| e | e |
| ia | ɯa |
| ie | ye |
| iia | ɯɯa |
| iie | yye |
| iiia | ɯɯɯa |
| iiie | yyy e |

The pattern continues for every nonnegative number of initial /i/ vowels. An engineer claims that exact normalization is possible with an ordinary deterministic finite-state device, because the input and output alphabets are both finite.

1. Let \(D=\{i^k a:k\ge 0\}\cup\{i^k e:k\ge 0\}\), and let \(f:D\to\{\text{i},\text{a},\text{e},\text{ɯ},\text{y}\}^*\) be the surface mapping specified in the Context.
2. A deterministic subsequential transducer consists of a finite state set \(Q\), an initial state, a deterministic transition function that reads one input symbol at a time and emits a fixed finite output string on each transition, and a fixed finite final output string associated with each state.
3. The transducer must compute \(f\) exactly for every member of \(D\); it may not revise output already emitted, use unbounded storage, or inspect unread input.
4. Outputs are compared as literal strings, so [ɯ] and [y] are distinct symbols.

Prove or disprove the engineer’s claim that such a deterministic subsequential transducer exists.",4 Language,40 Language,a very difficult,True,human_professional,math_reasoning,4 Language
q-corpus-1/q50_5/20,q50_5,20,"I'm the lead researcher on the new permanent gallery at our museum, and I've hit a genuine disagreement with our external academic advisor that I need help resolving before the panel copy goes to design in two weeks.

The disputed panel covers the Munich Agreement of September 1938. Our advisor wants the panel to carry a firm interpretive line: that Chamberlain's appeasement was a rational strategy of buying time for British rearmament rather than a naive delusion about Hitler's intentions. Our director thinks that framing is too generous to Chamberlain and will read to visitors as excusing the abandonment of Czechoslovakia. I've been asked to draft a memo that lays out the strongest case on each side so the curatorial board can decide what the panel actually says.

The claim to be argued for and against is this: ""Appeasement at Munich was primarily a calculated delay for rearmament, not a failure of judgment.""

For context, the claim leans partly on Chamberlain's own contemporary writing. In a letter to his sister Hilda in March 1938, after the Anschluss, he wrote: ""You have only to look at the map to see that nothing that we or anyone else could do could possibly save Czechoslovakia from being overrun by the Germans, if they wanted to do so... how horrible, fantastic, incredible it is that we should be digging trenches and trying on gas-masks here because of a quarrel in a far away country between people of whom we know nothing."" Our advisor reads this as cold strategic assessment of military reality; our director reads it as indifference dressed up as realism.

I need a single memo, no more than about 800 words, that does the following:

1. Presents the best-supported arguments FOR the claim, drawing on the state of British rearmament in 1938 (radar network, Spitfire production, the shadow factory scheme), the assessment of the Chiefs of Staff about the inability to defend Czechoslovakia, and the Dominion and French reluctance for war.

2. Presents the best-supported arguments AGAINST the claim, including the intelligence available about German weaknesses in autumn 1938, the Czech fortifications and Skoda arms works that fell to Germany, the unwillingness to explore a Soviet alliance or a firmer line before March 1938, and the possibility that Munich actually worsened Britain's strategic position by 1939.

3. Interprets the Chamberlain extract explicitly — does the letter support the ""rational delay"" reading, the ""naive delusion"" reading, or neither?

4. Closes with your own reasoned recommendation about which framing the panel should adopt, and be honest about where the historical evidence is genuinely contested rather than settled.

Please distinguish clearly between what the documentary record supports and what remains interpretive disagreement among historians — the board will notice if those are blurred. Deliverable is just the memo itself.",9 History,90 History,a difficult,False,human_professional,world_knowledge,9 History
q-corpus-1/q50_5/21,q50_5,21,"From Ray Bradbury's *Fahrenheit 451*: The firehouse keeps an eight-legged, insect-like machine called the Mechanical Hound. It sleeps in its kennel but can be chemically programmed to track a specific scent, and its proboscis injects a lethal dose of procaine or morphine. Captain Beatty uses it to intimidate Montag, and it is later deployed to hunt him through the city.

What is the Mechanical Hound in *Fahrenheit 451*, and what role does it play in the novel's society?",8 Literature,"80 Literature, rhetoric and criticism",a very simple,True,agent_structured,world_knowledge,8 Literature
q-corpus-1/q50_5/22,q50_5,22,"Hi team, in the coastal uplift study we're a bit puzzled by the numbers. The GPS is giving us 3 mm per year right now, but looking at the marine terrace which sits 120 meters above current sea level and dates back 120000 years when eustatic sea level was 50 meters below present, can you calculate what the average uplift rate must have been in mm per year based on the terrace.",5 Science,55 Earth sciences and geology,a,False,human_professional,math_reasoning,5 Science
q-corpus-1/q50_5/23,q50_5,23,"I'm trying to settle a dinner-table argument: in utilitarianism, what is the greatest happiness principle supposed to decide about the standard trolley case—should a bystander pull the lever, killing one person instead of five?",1 Philosophy and psychology,17 Ethics,a very simple,False,human_casual,world_knowledge,1 Philosophy and psychology
q-corpus-1/q50_5/24,q50_5,24,"request_id: GEO-417
task: compose
inputs:
- Calembert is an alluvial-island settlement in the tidal River Tess.
- The eastern branch lies between the island and the east bank; the western branch lies between the island and the west bank.
- At the provincial survey, the eastern branch was the main channel, so the boundary was drawn along its thalweg; the western branch was then a minor side channel.
- A flood avulsed the river into the western branch. The western branch is now deep and carries most flow; the eastern branch is silting and shallow at low tide, but still carries a permanent road bridge.
- The boundary follows the thalweg of the main channel, but after a sudden avulsion it remains in the former channel until both provinces approve a resurvey.
- The western branch has no bridge and is bordered by tidal marsh.
constraints: use only these inputs; keep a neutral gazetteer voice; do not invent additional facts.
ask: Compose a gazetteer paragraph explaining how the physical river change and the boundary rule can leave Calembert formally assigned to the west-bank province while its reliable land access is toward the east bank.",9 History,91 Geography,a difficult,False,agent_structured,creative_synthesis,9 History
q-corpus-1/q50_5/25,q50_5,25,"{
  ""task"":""explain"",
  ""focus"":""green light as rhetorical argument for the novel’s critique of the American Dream"",
  ""work"":""The Great Gatsby"",
  ""author"":""F. Scott Fitzgerald"",
  ""passage"":""He stretched out his arms toward the dark water in a curious way, and, far as I was from him, I could have sworn he was reaching toward a green light, trembling in the far distance across the water near the dock."",
  ""critical_method"":""symbolic analysis"",
  ""constraints"":{""max_words"":200}
}",8 Literature,"80 Literature, rhetoric and criticism",a simple,False,agent_structured,world_knowledge,8 Literature
q-corpus-1/q50_5/26,q50_5,26,"Available tools:
- list_upcoming_events(person_id: str [required], from_date: str [required], to_date: str [required]) -> list of event dicts
  List upcoming events for a person within a date range.
- query_database(sql: str [required], database: str [required], max_rows: int = 100) -> dict with rows and column_names
  Execute a SQL query and return matching rows.
- create_incident(title: str [required], severity: str, enum: ['sev1','sev2','sev3','sev4'] [required], description: str [required], affected_services: list[str] [required]) -> dict with incident_id
  Create an incident report.
- count_records(table: str [required], database: str [required], where: str = None) -> int
  Count records matching optional filter criteria.

Investigate a possible spike in failed checkouts. In `payments_db`, count rows in the `transactions` table where `status = 'failed'` and `created_at >= '2025-01-15'`. I need the result as a single integer count, not a row dump.

If that count exceeds 50, open an incident with severity `sev2`, title `""Elevated checkout failures""`, affected services `[""checkout-api"", ""payments-gateway""]`, and a description that states the exact count returned by the first step. Do not create the incident if the count is 50 or below — in that case just report the number.

Both steps must run in order; the second depends on the value produced by the first.",A Agentic tool use,tool_selection,a difficult,True,agent_structured,planning_agentic,A Agentic tool use
q-corpus-1/q50_5/27,q50_5,27,"I'm running this PID loop on an embedded controller for a small resistive heater. It compiles and runs, but during commissioning the block keeps tripping the over-temp interlock a few seconds after startup, then oscillates instead of settling. I already checked the thermocouple wiring and the ADC scaling — both are fine.

Given conditions:
1. Sample period is fixed at 100 ms (called from a hardware timer ISR).
2. Setpoint is 85.0 °C, starting from ambient 22.0 °C.
3. Heater output `pwm_duty` must stay within 0–255.
4. Hard safety interlock trips if measured temp exceeds 95.0 °C.
5. Gains: Kp = 6.0, Ki = 0.8, Kd = 3.5.
6. Actuator has real physical lag: full duty change takes ~4 s to show up in the temperature reading.

```c
typedef struct {
    float kp, ki, kd;
    float integral;
    float prev_error;
} pid_t;

int pid_update(pid_t *p, float setpoint, float measured, float dt) {
    float error = setpoint - measured;

    p->integral += error * dt;
    float derivative = (error - p->prev_error) / dt;

    float output = p->kp * error
                 + p->ki * p->integral
                 + p->kd * derivative;

    p->prev_error = error;

    int duty = (int)output;
    if (duty > 255) duty = 255;
    if (duty < 0)   duty = 0;

    return duty;
}
```

Given the actuator lag in item 6, explain why this controller drives the temperature past the 95 °C interlock and then oscillates, and rewrite `pid_update` so the integral term stops accumulating once `output` is already saturated at 0 or 255, without changing the function signature or the gains in item 5.",6 Technology,62 Engineering,a difficult,True,human_casual,coding,6 Technology
q-corpus-1/q50_5/28,q50_5,28,"I'm cross-referencing our demographic notes with the text of Article I, Section 2 of the 1787 U.S. Constitution, which directs that representation be apportioned by adding ""to the whole Number of free Persons... three fifths of all other Persons."" 

If a 1790 county return lists 25,000 free inhabitants and 10,000 enslaved individuals, what is the exact representative population figure for that county under this constitutional rule?",9 History,90 History,a very simple,False,human_casual,math_reasoning,9 History
q-corpus-1/q50_5/29,q50_5,29,"Consider the processive stepping of kinesin-1 on a microtubule protofilament. The motor advances via a hand-over-hand mechanism with a structural step size $d = 8.2\text{ nm}$ per ATP hydrolyzed. The mechanochemical cycle is modeled as a continuous-time Markov chain with four principal states per head:  
1. **Empty (E)**: Nucleotide-free, tight microtubule binding.  
2. **ATP-bound (T)**: ATP bound, neck linker docked (power stroke primed).  
3. **ADP·Pi-bound (D\*)**: Post-hydrolysis, loose binding.  
4. **ADP-bound (D)**: ADP bound, weak binding, diffusive search.

Transition rates (s$^{-1}$) at zero external load ($F=0$) and saturating ATP are:  
$k_{\text{hyd}} = 120$ (E+ATP $\to$ T), $k_{\text{pi}} = 200$ (T $\to$ D\*), $k_{\text{adp}} = 50$ (D\* $\to$ D), $k_{\text{on}} = 2.0 \times 10^6 \text{ M}^{-1}\text{s}^{-1}$ (D+ATP $\to$ T).  
The rear head detaches only in state D with rate $k_{\text{off}} = 5000$. The forward step occurs upon neck linker docking (transition E+ATP $\to$ T), biased by the free energy of ATP binding $\Delta G_{\text{bind}}$. Load $F$ acts as a tilting force on the step transition, modifying the forward ($k_+$) and reverse ($k_-$) rates via a load distribution factor $\theta = 0.5$: $k_\pm(F) = k_\pm(0) \exp(\mp \theta F d / k_B T)$.

**Task:**  
1. Derive the analytical expression for the steady-state velocity $v(F, [\text{ATP}])$ as a function of load and ATP concentration using the King-Altman method or flux-balance on the 4-state cycle, assuming rapid equilibrium for ATP binding relative to hydrolysis.  
2. Prove that the stall force $F_s$ satisfies $F_s d = \Delta G_{\text{ATP}} + k_B T \ln([\text{ADP}][\text{Pi}]/[\text{ATP}])$ exactly, independent of the kinetic rates $k_{\text{hyd}}, k_{\text{pi}}, k_{\text{adp}}$, provided detailed balance holds for the chemical cycle.  
3. Compute the randomness parameter $r = \lim_{t \to \infty} \text{Var}(x(t)) / \langle x(t) \rangle$ at $F=0$ and $[\text{ATP}] = 1 \text{ mM}$ using the cycle completion time distribution. Determine if the motor operates in a Poissonian ($r=1$) or sub-Poissonian ($r<1$) regime given the stated rates.  
Use $k_B T = 4.1 \text{ pN}\cdot\text{nm}$, $\Delta G_{\text{ATP}} = -20 k_B T$. Show all derivations.",5 Science,57 Biology,a difficult,False,agent_structured,math_reasoning,5 Science
q-corpus-1/q50_5/30,q50_5,30,"Finite-state morphological analyzer for Turkish nominal inflection. Stem = `kitap` with final vowel class back unrounded {a,ı}. Suffix chain order is fixed: [NUMBER] [POSSESSOR] [CASE] with each slot optional, at most one suffix per slot, and ∅ as the null exponent.

Inventory with underlying vowel slot V and consonant material:
NUMBER: ∅, -lAr
POSSESSOR: ∅, -Im, -In, -I, -ImIz, -InIz, -IlerI
CASE: ∅, -DA, -DAn, -yA

Vowel harmony mapping for V:
stem back unrounded → a in suffixes, stem back rounded → o, stem front unrounded → e, stem front rounded → ö.
Consonant assimilation:
-lAr surfaces as -lar after a back unrounded stem, -ler after front unrounded, -lar after back rounded with rounding neutralized, -ler after front rounded.
Buffer consonant d is inserted before vowel-initial suffixes when stem ends in p,t,k,c,f,s,ş,h, and assimilates to the stem-final consonant: p→p, t→t, k→k, ç→ç, f→f, s→s, ş→ş, h→h. For `kitap` the stem-final consonant is p, so the buffer surfaces as p.

Allomorphy is deterministic given stem vowel class and final consonant.

Given the specification above, compute the number of distinct surface forms for the stem `kitap`, counting the bare stem as one form, and provide the closed-form count for an arbitrary stem with a fixed vowel class and no stem-final consonant alternation other than the buffer rule described.",4 Language,40 Language,a difficult,True,human_professional,math_reasoning,4 Language
q-corpus-1/q50_5/31,q50_5,31,"Compose a first-person diary entry dated June 20 1789 by a Paris artisan who joined the locked-out Third Estate at the tennis court, showing how the royal treasury crisis and refusal of voting reform pushed him to support an oath binding the delegates together.",9 History,90 History,a simple,False,agent_structured,creative_synthesis,9 History
q-corpus-1/q50_5/32,q50_5,32,"A borehole drilled in the continental shelf offshore Mauritania intersected a continuously cored section with the following stratigraphic control:

- **0–220 m below seafloor (mbsf):** *Globigerinoides ruber* (pink variant), last occurrence 0.12 Ma, present throughout.
- **220–340 mbsf:** *Globorotalia truncatulinoides* (sinistral), first occurrence 1.77 Ma, present throughout.
- **340–410 mbsf:** *Globorotalia margaritae*, last occurrence 3.85 Ma, present throughout.
- **410 mbsf:** acoustic basement (basement conglomerate).

The drilling team claims that the interval from 220–340 mbsf represents steady pelagic carbonate accumulation at roughly 7 cm/kyr, consistent with regional deep-water sedimentation rates. The operations geologist disputes this, arguing the rate is materially lower and that the interval is condensed.

Calculate the linear sedimentation rate (cm/kyr) for the 220–340 mbsf interval using the biostratigraphic age picks given. Then determine whether the team's claim of ~7 cm/kyr is supported.",5 Science,55 Earth sciences and geology,a simple,False,human_professional,math_reasoning,5 Science
q-corpus-1/q50_5/33,q50_5,33,"The dataset comprises 8 observation pairs from the grammaticalization of *sǝj* 'to exist' into a definiteness marker in a Hmong-Mien dialect. Each entry lists the year of observation, the surface form with tonal specification, a morpheme-by-morpheme gloss, an English translation, and the values of two binary features: [±high tone] on the host syllable (H/L) and [±human] of the following noun phrase (+H/–H).

Observation A: Year 1; surface *sǝj* H +H; gloss *sǝj* txoj; translation 'there is a book'; features reflect existential clause.
Observation B: Year 3; surface *sǝj* L –H; gloss *sǝj* txoj; translation 'there is a stone'; features reflect existential clause.
Observation C: Year 5; surface *sì* H +H; gloss *sì* txoj; translation 'the book'; definite marker appears.
Observation D: Year 5; surface *sì* L –H; gloss *sì* txoj; translation 'the stone'; definite marker appears.
Observation E: Year 7; surface *sì* H +H; gloss *sì‑mə* txoj; translation 'the book'; suffixal form emerges.
Observation F: Year 7; surface *sì* L +H; gloss absent; translation 'the book' (no suffix) despite high tone and human NP.
Observation G: Year 9; surface *sì* L –H; gloss *sì* txoj; translation 'the stone'; suffixal form appears.
Observation H: Year 10; surface *sì* H –H; gloss absent; translation 'the book' (no suffix) despite high tone and non‑human NP.

Based exclusively on observations A–H, at what grammaticalization stage did the [±human] feature of the complement become the decisive constraint on suffix distribution, and which single phonological environment (high or low tone on the host) appears to have been overridden by this semantic constraint? Justify your conclusion using only the data provided.",4 Language,40 Language,a very difficult,False,human_casual,world_knowledge,4 Language
q-corpus-1/q50_5/34,q50_5,34,"I'm writing up an incident report for our sales floor and need to document Mark's recent behavior using the right psychological framework. During our last team meeting, Mark strongly advocated for open collaboration and shared success. But yesterday, we caught him hoarding a new client list so he could hit his individual sales bonus, directly contradicting his stated values. When his manager confronted him about the discrepancy, Mark immediately became highly defensive and started justifying his actions by claiming his peers ""don't work as hard as I do anyway."" 

How would Leon Festinger's theory of cognitive dissonance explain Mark's sudden defensiveness and his attempt to justify hoarding the client list?",1 Philosophy and psychology,15 Psychology,a simple,False,human_professional,world_knowledge,1 Philosophy and psychology
q-corpus-1/q50_5/35,q50_5,35,"The regional transmission organization (RTO) currently operates an energy-only wholesale electricity market supplemented by an Operating Reserve Demand Curve (ORDC) with a Value of Lost Load (VOLL) administrative cap of $5,000/MWh. Due to the rapid addition of zero-marginal-cost renewable generation (projected to reach 48% of nameplate capacity by 2028), wholesale spot energy prices have exhibited prolonged periods of near-zero clearing, interspersed with localized, short-duration price spikes. 

Thermal merchant generators argue that suppressed average spot prices create an acute ""missing money"" problem that impedes the financing of fast-ramping open-cycle gas turbines (OCGT) and multi-hour battery storage systems required for reliability during net-load ramp events.

In response, the RTO Market Monitor has put forward a proposal to transition from the ORDC energy-only design to a 3-Year Forward Centralized Capacity Market (CCM) using a downward-sloping demand curve anchored to the Net Cost of New Entry (Net CONE), coupled with two-settlement capacity performance obligations.

Key market parameters under evaluation:
1. Current Reserve Margin: 11.2% (Target: 15.0%).
2. Historical Frequency of ORDC Scarcity Pricing Activation: 14 hours/year over the last 36 months.
3. Estimated Annualized Gross CONE for 4-hour Li-ion Battery Storage: $128/kW-year.
4. Average Energy and Ancillary Service (E&AS) Offset captured by flexible assets: $42/kW-year.
5. Capital Cost Differential: Merchant financing cost of debt is estimated at 8.5% under the energy-only framework versus 6.1% under 3-year capacity supply obligations.

Analyze the validity of the following claim: ""Replacing the ORDC scarcity-pricing mechanism with a 3-Year Forward Centralized Capacity Market will lower aggregate consumer procurement costs and resolve the missing-money problem for flexible peaking assets without causing structural capacity over-procurement or distorting operational dispatch incentives.""

Provide the strongest economic arguments supporting this claim (drawing on the Cramton–Stoft capacity mechanism framework and financing risk premia) and the strongest counterarguments against it (drawing on Hogan’s scarcity pricing theory, administrative Net CONE estimation biases, and potential suppression of demand-side response incentives).",3 Social Science,33 Economics,a difficult,True,human_professional,world_knowledge,3 Social Science
q-corpus-1/q50_5/36,q50_5,36,"{
  ""task"": ""compare"",
  ""item1"": {
    ""match"": ""Super Bowl LI"",
    ""teams"": ""Patriots vs Falcons"",
    ""segment"": ""final 5 minutes""
  },
  ""item2"": {
    ""match"": ""Super Bowl XXXIV"",
    ""teams"": ""St. Louis Rams vs Tennessee Titans"",
    ""segment"": ""final 5 minutes""
  },
  ""focus"": ""score differential evolution, offensive tempo, and defensive adjustments; assess how each late‑game strategy impacted scoring efficiency and momentum shift"",
  ""constraints"": {
    ""max_words"": 300,
    ""tone"": ""analytical"",
    ""format"": ""bullet points for each game followed by a synthesis""
  },
  ""output_requirement"": ""highlight non‑obvious strategic divergences despite comparable game states""
}",7 Arts & recreation,"79 Sports, games and entertainment",a,False,agent_structured,world_knowledge,7 Arts & recreation
q-corpus-1/q50_5/37,q50_5,37,"Passage 1 (T.S. Eliot, ""Hamlet and His Problems"", 1919): ""The only way of expressing emotion in the form of art is by finding an 'objective correlative'; in other words, a set of objects, a situation, a chain of events which shall be the formula of that particular emotion; such that when the external facts, which must terminate in sensory experience, are given, the emotion is immediately evoked.""

Passage 2 (John Ruskin, Modern Painters, Vol. III, Part IV, Chapter 12): ""The pathetic fallacy is a falseness in the imagination of the painter or poet, which makes him attribute to inanimate objects the emotions or sensations of living creatures, not out of a true sympathy, but from a morbid instinct of indulgence in his own feelings.""

I'm finalizing the headnote for our graduate anthology of literary criticism and these two passages are creating a classification snag: Eliot's objective correlative and Ruskin's pathetic fallacy seem to pull in opposite directions on whether external objects should carry emotion. Please compare the two concepts as defined in the excerpts above, analysing how each theorises the link between inanimate objects and human feeling in an existing work, and indicate which one presupposes a healthier artistic process.",8 Literature,"80 Literature, rhetoric and criticism",a,True,human_professional,world_knowledge,8 Literature
q-corpus-1/q50_5/38,q50_5,38,"How does the learned helplessness mechanism link the clinical observation that a depressed client who after several unanswered job applications now refuses interviews saying ""nothing I do matters"" to Seligman's finding that dogs given inescapable shock later failed to escape when shock became escapable, and what core cognitive distortion connects this resignation to the original animal result?",1 Philosophy and psychology,15 Psychology,a simple,False,agent_structured,world_knowledge,1 Philosophy and psychology
q-corpus-1/q50_5/39,q50_5,39,"Available tools:
- compute_correlation(dataset: str [required], column_a: str [required], column_b: str [required]) -> dict with correlation_coefficient and p_value
  Compute correlation between two metric columns.
- create_table(table: str [required], columns: list[dict] [required], database: str [required]) -> bool
  Create a new table with specified columns.
- get_table_schema(table: str [required], database: str [required]) -> list of column definitions
  Get column names, types, and constraints for a table.
- schedule_meeting(title: str [required], attendees: list[str] [required], start_time: str [required], duration_min: int [required], location: str = '') -> dict with meeting_id
  Schedule a meeting with attendees.
- update_record(table: str [required], record_id: str [required], fields: dict [required], database: str [required]) -> bool
  Update fields of an existing record by ID.

We need to process the telemetry audit for turbine anomaly REF-9904. Please plan and execute the required tool calls following this operational protocol:

First, inspect the schema for the staging data definitions: query the table `telemetry_timeseries_v3` inside the `propulsion_analytics_db` catalog to verify the active metric fields. 

Next, run a correlation calculation between the two target metric variables for dataset `flight_run_2024_oct_89`: the rotor vibration metric (`rotor_vib_axial_g`) and the turbine inlet temperature (`tit_inlet_kelvin`), ignoring the ambient baseline columns (`temp_ambient_c` and `rotor_vib_radial_g`) that were initially flagged by mistake.

Take the resulting statistical values and update our primary audit tracking entry. The target record ID is `REF-9904`, located in the `incident_ledger` table within the `compliance_audit_prod` database. Populate the record's fields dictionary with:
- `corr_r`: the calculated correlation coefficient
- `significance_p`: the calculated p-value
- `auditor_email`: `j.reynolds@aerotech-ops.net`
- `status`: `EVALUATED`

Finally, organize the mandatory review meeting. The debrief session must be titled ""Telemetry Audit Debrief - REF-9904"", set for a duration of 45 minutes, with the start time scheduled at `2024-10-31T15:30:00Z` in conference room ""Avionics Lab 4B"". Gather all required participants from the incident chain: the reporting engineer (j.reynolds@aerotech-ops.net), the lead propulsion specialist (k.sato@aerotech-ops.net), and the quality assurance liaison (qa-escalations@aerotech-ops.net).",A Agentic tool use,parameter_construction,a very difficult,True,human_casual,planning_agentic,A Agentic tool use
q-corpus-1/q50_5/40,q50_5,40,"I’m trying to help a junior manuscript reviewer stop labeling every first-person narrator as “unreliable.” Could you draft a short, collegial email that explains the narratological concept of unreliable narration? Use this opening fragment from Poe’s “The Tell-Tale Heart” as your illustrative example: “TRUE! —nervous —very, very dreadfully nervous I had been and am; but why will you say that I am mad? The disease had sharpened my senses —not destroyed —not dulled them. Above all was the sense of hearing acute. I heard all things in the heaven and in the earth. I heard many things in hell. How, then, am I mad?” In the email, identify the specific textual cues that signal genuine unreliability rather than simple subjectivity, and remind the recipient how to apply that distinction in future reader’s reports. Keep the tone practical and the total length under 180 words.",8 Literature,"80 Literature, rhetoric and criticism",a simple,False,human_professional,creative_synthesis,8 Literature
q-corpus-1/q50_5/41,q50_5,41,"rewrite_as_brief_explanation
work: Edgar Allan Poe, ""The Tell-Tale Heart""
method: psychoanalytic criticism
idea: projection
given_notes: narrator denies motive, calls the eye ""vulture-like,"" insists he loves the man, then hears a heartbeat no one else confirms
ask: Compose 2–3 sentences for a gallery label that explains how this reading treats the eye and heartbeat as projections of guilt rather than facts in the story. Keep the narrator's actions to one short clause. Use plain present tense.",8 Literature,"80 Literature, rhetoric and criticism",a simple,False,agent_structured,creative_synthesis,8 Literature
q-corpus-1/q50_5/42,q50_5,42,"1. A shipped text-to-speech morphophonology module for the constructed agglutinative language Lëz (typologically based on Uralic vowel harmony) began emitting unnatural suffix vowels on loanword tokens after a weighted OT update, creating a regression where non-round suffixes unexpectedly resist backness harmony while round suffixes harmonize.
2. Field transcriptions confirm a systematic split: a [+back] root with a [-back,+round] suffix surfaces with the suffix backed to [+back,+round], and a [-back] root with a [+back,+round] suffix surfaces with the suffix fronted to [-back,+round]; in both cases the round feature is preserved and harmony applies only when the suffix is [+round].
3. The module implements a tropical-semiring OT grammar with three ranked constraints: C1 = *[+back,-round]/Suffix (weight 5.0), C2 = Agree(back) (weight 3.0), C3 = Max(round) (weight 1.0), where C1 dominates C2 dominates C3.
4. Constraint violation definitions: Agree(back) assigns one violation per suffix vowel whose [back] value differs from the root; *[+back,-round]/Suffix assigns one violation per suffix vowel that is [+back,-round]; Max(round) assigns one violation per lost [+round] specification on the suffix.
5. The candidate set for each underlying representation (UR) is restricted to the observed surface form (SR) and the strictly harmonic counterpart that matches the suffix [back] value to the root and preserves any [+round] specification.
6. The following four representative UR→SR pairs (feature values given as [back,round]) were extracted from the regression logs:

| ID | UR (root, suffix) | SR (root, suffix) | Harmonic counterpart (root, suffix) |
|----|-------------------|-------------------|-------------------------------------|
| A | [+back,-round], [-back,-round] | [+back,-round], [-back,-round] | [+back,-round], [+back,-round] |
| B | [-back,-round], [+back,-round] | [-back,-round], [-back,-round] | [-back,-round], [-back,-round] |
| C | [+back,-round], [-back,+round] | [+back,-round], [+back,+round] | [+back,-round], [+back,+round] |
| D | [-back,-round], [+back,+round] | [-back,-round], [-back,+round] | [-back,-round], [-back,+round] |

Compute the total weighted violation score (5.0·viol(C1) + 3.0·viol(C2) + 1.0·viol(C3)) for the observed SR and for the harmonic counterpart in each of A–D, and on that basis prove that the given constraint ranking selects the observed SR as optimal in every case, or disprove it by naming the case(s) where the harmonic counterpart achieves a strictly lower total score.",4 Language,40 Language,a very difficult,False,human_professional,math_reasoning,4 Language
q-corpus-1/q50_5/43,q50_5,43,"> Gatsby believed in the green light, the orgastic future that year by year recedes before us.  
>   
> So we beat on, boats against the current, borne back ceaselessly into the past.

Provide a detailed, scholarly argumentation both supporting and contesting the following claim: the final passage of *The Great Gatsby* functions as a deliberate rhetorical critique of the American Dream, employing its imagery and syntax to undermine the dream’s promise of upward mobility and progress rather than merely serving as a nostalgic lament of personal loss. In your response, reference specific linguistic features, thematic resonances, and relevant critical perspectives that can be marshaled on either side of the debate. Cite at least two distinct scholarly interpretations (published before 2024) that exemplify each position, and conclude with a brief assessment of which side of the argument appears more persuasive based on the textual evidence.",8 Literature,"80 Literature, rhetoric and criticism",a very difficult,True,human_professional,world_knowledge,8 Literature
q-corpus-1/q50_5/44,q50_5,44,"Hey — we inherited the WQRX community-radio archive after the station folded in 2019. Roughly 12k born-digital files. The NAS died; all we have is a recovery dump (filenames, mtimes), a couple of leftover .mbox extracts, and a volunteer CSV of on-air logs. Original folder hierarchy is gone.

We've stopped arguing in meetings and agreed to treat these as given:

H1. Every record is created by exactly one agent (a DJ, the engineer, the station manager, or ""station"" for automated playout). Agent is recoverable only from filename tokens, email From, or the log's host field — do not invent agents.

H2. If two records share an agent and their mtimes fall inside the same contiguous logged shift, they belonged to the same original file unit.

H3. Series are activities, not formats: Programming, Engineering, Governance, Correspondence. Activity follows agent+shift type, except Governance, which is only meeting minutes and board emails (detect from kind/subject), even if they land inside a broadcast shift.

H4. The broadcast log is the evidential spine: any arrangement that splits records from the same logged shift *and* same agent across different file units is wrong. Clock skew under 15 minutes still counts as inside the shift; larger disagreement means the record is not in that shift.

H5. Fonds is the station (WQRX). Item = one record. File = the reconstructed original aggregation. Unassigned records go to a series ""Unplaced"" with one file per record; they must not be silently dropped.

records.json and shifts.json are below (40-odd rows collapsed to a working subset). Write a single Python 3 script that reads those two structures (inline is fine), assigns each record a DACS-style identifier fonds/series/file/item consistent with H1–H5, reconstructs file-level aggregations from the shift windows, and prints (a) a nested finding-aid tree and (b) a comment block that walks the assignments and shows that no file unit violates H4. Don't invent agents, shifts, or series names that aren't licensed by the hypotheses and the data.

records = [
  {""id"":""r01"",""filename"":""playlist_dawn_20180312.m3u"",""mtime"":""2018-03-12T06:14:02Z"",""kind"":""playlist"",""email_from"":null,""subject"":null},
  {""id"":""r02"",""filename"":""maria_show_20180312.wav"",""mtime"":""2018-03-12T06:58:11Z"",""kind"":""audio"",""email_from"":null,""subject"":null},
  {""id"":""r03"",""filename"":""inbox.mbox"",""mtime"":""2018-03-12T07:02:44Z"",""kind"":""email"",""email_from"":""maria@wqrx.org"",""subject"":""re: board packet""},
  {""id"":""r04"",""filename"":""board_minutes_20180312.pdf"",""mtime"":""2018-03-12T19:40:00Z"",""kind"":""minutes"",""email_from"":null,""subject"":null},
  {""id"":""r05"",""filename"":""helen_to_board.eml"",""mtime"":""2018-03-12T19:12:18Z"",""kind"":""email"",""email_from"":""helen@wqrx.org"",""subject"":""agenda for tonight""},
  {""id"":""r06"",""filename"":""kole_nightmix_20180312.wav"",""mtime"":""2018-03-13T01:07:33Z"",""kind"":""audio"",""email_from"":null,""subject"":null},
  {""id"":""r07"",""filename"":""kole_tracklist.txt"",""mtime"":""2018-03-13T02:51:02Z"",""kind"":""playlist"",""email_from"":null,""subject"":null},
  {""id"":""r08"",""filename"":""tx_log_patel.csv"",""mtime"":""2018-03-13T04:22:00Z"",""kind"":""log"",""email_from"":null,""subject"":null},
  {""id"":""r09"",""filename"":""patel_notes.md"",""mtime"":""2018-03-13T04:28:41Z"",""kind"":""notes"",""email_from"":null,""subject"":null},
  {""id"":""r10"",""filename"":""autoplay_overnight.m3u"",""mtime"":""2018-03-13T05:01:09Z"",""kind"":""playlist"",""email_from"":null,""subject"":null},
  {""id"":""r11"",""filename"":""maria_show_20180313.wav"",""mtime"":""2018-03-13T06:47:55Z"",""kind"":""audio"",""email_from"":null,""subject"":null},
  {""id"":""r12"",""filename"":""unknown_dump.bin"",""mtime"":""2018-03-14T15:03:12Z"",""kind"":""unknown"",""email_from"":null,""subject"":null},
  {""id"":""r13"",""filename"":""kole_to_maria.eml"",""mtime"":""2018-03-12T23:58:01Z"",""kind"":""email"",""email_from"":""kole@wqrx.org"",""subject"":""can you cover Thursday""},
  {""id"":""r14"",""filename"":""studio_photo.jpg"",""mtime"":""2018-03-12T06:20:00Z"",""kind"":""photo"",""email_from"":null,""subject"":null},
  {""id"":""r15"",""filename"":""helen_to_fcc.eml"",""mtime"":""2018-03-14T11:11:11Z"",""kind"":""email"",""email_from"":""helen@wqrx.org"",""subject"":""translator construction permit""}
]

shifts = [
  {""id"":""S-A"",""start"":""2018-03-12T06:00:00Z"",""end"":""2018-03-12T10:00:00Z"",""host"":""dj.maria"",""activity"":""Programming""},
  {""id"":""S-B"",""start"":""2018-03-12T19:00:00Z"",""end"":""2018-03-12T21:00:00Z"",""host"":""mgr.helen"",""activity"":""Governance""},
  {""id"":""S-C"",""start"":""2018-03-13T00:00:00Z"",""end"":""2018-03-13T03:00:00Z"",""host"":""dj.kole"",""activity"":""Programming""},
  {""id"":""S-D"",""start"":""2018-03-13T04:00:00Z"",""end"":""2018-03-13T05:00:00Z"",""host"":""eng.patel"",""activity"":""Engineering""},
  {""id"":""S-E"",""start"":""2018-03-13T05:00:00Z"",""end"":""2018-03-13T06:00:00Z"",""host"":""station"",""activity"":""Programming""},
  {""id"":""S-F"",""start"":""2018-03-13T06:00:00Z"",""end"":""2018-03-13T10:00:00Z"",""host"":""dj.maria"",""activity"":""Programming""}
]

Filename tokens: maria → dj.maria, kole → dj.kole, patel → eng.patel, helen → mgr.helen, autoplay/playlist_dawn with no person token → station. Email From maps the same way. r14 has maria's shift window but no agent token — leave it Unplaced rather than guessing. r13 sits 2 minutes before S-C; treat skew per H4. Correspondence is only person-to-person email that is not board/governance.","0 Computer science, information, and general works",02 Library and information sciences,a very difficult,False,human_casual,coding,"0 Computer science, information, and general works"
q-corpus-1/q50_5/45,q50_5,45,"Hi team,

In the latest iteration of our allocation model each hierarchical level i is associated with the i‑th triangular number  

\[
T_i=\frac{i(i+1)}{2}.
\]

The overall risk metric we need to compute involves the infinite series  

\[
S=\sum_{i=1}^{\infty}\frac{1}{T_i^{2}}.
\]

Using the definition of \(T_i\) we can rewrite the summand as  

\[
\frac{1}{T_i^{2}}=\frac{4}{i^{2}(i+1)^{2}}.
\]

A quick algebraic step yields the partial‑fraction identity  

\[
\frac{4}{i^{2}(i+1)^{2}}
=4\Bigl(\frac{1}{i^{2}}+\frac{1}{(i+1)^{2}}-\frac{2}{i(i+1)}\Bigr).
\]

The first two terms will generate a zeta‑type sum, while the last term telescopes.  

Could you provide the exact closed‑form value of \(S\) (expressed in terms of \(\pi\) or other elementary constants) and a concise justification that follows from the decomposition above?",5 Science,51 Mathematics,a,False,human_professional,math_reasoning,5 Science
q-corpus-1/q50_5/46,q50_5,46,"Hi — I'm putting together a memo on the proposed $22.50/cell ad valorem tariff (15% of world price) on imported lithium-ion battery cells and need the welfare math worked out properly. Here's what I'm working with.

Domestic demand is Qd = 800 − 2P (millions of cells/year). Domestic supply is Qs = 100 + P. The world price is $150/cell, and at that price imports fill the residual demand. Treat this as a small country — our market is negligible relative to global production.

Using the standard partial-equilibrium tariff framework, derive the changes in consumer surplus, domestic producer surplus, government tariff revenue, and deadweight loss relative to free trade. Start from the demand and supply equations and show the algebra. The key thing I need answered: under these parameters, does tariff revenue exceed the combined production and consumption deadweight losses, or is this a net welfare loss? Give me the dollar figure for the gap either way.

I want the derivation shown, not just final numbers.",3 Social Science,33 Economics,a,False,human_casual,math_reasoning,3 Social Science
q-corpus-1/q50_5/47,q50_5,47,"I have a Rust program using Tokio that intermittently hangs or panics at runtime. The code is:

```rust
use std::sync::{Arc, Mutex};
use tokio::time::{sleep, Duration};

struct Counter {
    value: i32,
}

#[tokio::main]
async fn main() {
    let counter = Arc::new(Mutex::new(Counter { value: 0 }));
    let c1 = counter.clone();
    let c2 = counter.clone();

    let t1 = tokio::spawn(async move {
        let mut guard = c1.lock().unwrap();
        guard.value += 1;
        sleep(Duration::from_millis(100)).await;
        guard.value += 2;
    });

    let t2 = tokio::spawn(async move {
        sleep(Duration::from_millis(50)).await;
        let mut guard = c2.lock().unwrap();
        guard.value *= 3;
    });

    let _ = tokio::join!(t1, t2);
    println!(""Final: {}"", counter.lock().unwrap().value);
}
```

Observed behavior: the program frequently deadlocks (the process never terminates) or panics with “cannot lock mutex more than once at a time”.

**Tasks**
1. Identify the root cause of the deadlock/panic in this async context.
2. Rewrite the snippet so it runs correctly without altering the intended arithmetic (increment by 1, pause, add 2, while a concurrent task multiplies by 3 after a shorter pause). Use only Tokio‑compatible primitives.
3. Provide a brief explanation (a few sentences) of how the interaction of three facts—(a) `std::sync::Mutex` is a blocking lock, (b) an `.await` occurs while the mutex guard is held, and (c) Tokio’s cooperative scheduler—produces the observed failure.

Return the fixed code followed by the concise explanation.","0 Computer science, information, and general works","00 Computer science, knowledge, and systems",a,False,agent_structured,coding,"0 Computer science, information, and general works"
q-corpus-1/q50_5/48,q50_5,48,"I’m debugging a medication reconciliation helper that compares FHIR R4 `MedicationRequest` and `MedicationStatement` entries for one patient. The intended comparison is: ignore any entry with status `entered-in-error`; group remaining entries by `""{system}|{code}""`; for each group, return `source=""request""` if at least one request remains, otherwise `source=""statement""`. Set `active` to true if any remaining request in the group has status `active` or `on-hold`, or any remaining statement has status `active`. In particular, a stopped request must not make the grouped record inactive when an active statement exists. A missing `system` or missing `code` should be skipped, not treated as a wildcard.

Please rewrite `merge_medication_entries` so it returns a list of dicts with `key`, `display`, `source`, and `active`. Use the display from the selected source; if absent, use the code portion of the key.

```python
def merge_medication_entries(requests, statements):
    out = []
    for req in requests:
        coding = req.get(""coding"") or {}
        key = f""{coding.get('system')}|{coding.get('code')}""
        out.append({
            ""key"": key,
            ""display"": req.get(""display"") or coding.get(""code""),
            ""source"": ""request"",
            ""active"": req.get(""status"") == ""active""
        })
    for stmt in statements:
        coding = stmt.get(""coding"") or {}
        key = f""{coding.get('system')}|{coding.get('code')}""
        out.append({
            ""key"": key,
            ""display"": stmt.get(""display"") or coding.get(""code""),
            ""source"": ""statement"",
            ""active"": stmt.get(""status"") == ""active""
        })
    return out
```

Deliverable: the corrected function only.",6 Technology,61 Medicine and health,a,False,human_professional,coding,6 Technology
q-corpus-1/q50_5/49,q50_5,49,"Hey, quick one — in this Python snippet, what does the walrus operator `:=` actually do here?

```python
data = [1, 2, 3, 4, 5, 6, 7, 8]
results = []
i = 0
while (n := data[i]) < 7:
    results.append(n)
    i += 1
print(results)
```

I get the general idea of assignment expressions but I can't tell offhand what `results` ends up holding when this runs, or why someone would bother writing the loop condition this way instead of just doing `data[i] < 7` directly. Can you explain what's happening line by line and what the final output is?","0 Computer science, information, and general works","00 Computer science, knowledge, and systems",a very simple,False,human_professional,coding,"0 Computer science, information, and general works"
q-corpus-1/q50_5/50,q50_5,50,"Hey, need a quick hand for a display we're putting up at work - we're doing a timeline on D-Day (June 6, 1944) and how it led directly to the Liberation of Paris in August 1944. For the caption we want to say how long ago D-Day was as of June 6, 2024. Can you calculate and prove the math to show exactly how many years that is?",9 History,90 History,a very simple,False,human_professional,math_reasoning,9 History
q-corpus-1/q100_6/1,q100_6,1,"1. A startup has trained a conversational avatar on the private messages, voice notes, and photographs of Mara, who died unexpectedly at 34.
2. Mara never consented to posthumous AI reconstruction; her spouse approved the pilot, while her sister—whose intimate messages are included—objected.
3. The company says the avatar helps grieving users and will delete any material on request, but it retains commercial rights to improve its system and market subscriptions.
4. Early users report comfort, yet some begin treating the avatar as Mara’s continuing presence and delay decisions about funeral arrangements, inheritance, and parenting.
5. The draft ethics-board statement below says only: “Because the tool provides comfort and the spouse consented, deployment should continue with transparent disclosures.”
6. The board must decide whether relational benefits can justify creating a persuasive imitation without the deceased person’s authorization, and must address who controls another person’s traces, how grief may be monetized, and what obligations are owed to living contributors.

Rewrite the draft as a clear public-facing ethics-board statement that takes a definite position, identifies what is morally at stake, and gives a defensible policy for this deployment.",1 Philosophy and psychology,17 Ethics,a difficult,False,agent_structured,creative_synthesis,1 Philosophy and psychology
q-corpus-1/q100_6/2,q100_6,2,"Available tools:
- refresh_auth_token(token_url: str [required], refresh_token: str [required]) -> dict with access_token and expires_in
  Refresh an expired authentication token.
- create_recurring_event(title: str [required], person_id: str [required], start_time: str [required], duration_min: int [required], recurrence: str, enum: ['daily','weekly','biweekly','monthly'] [required], end_date: str = None) -> dict with event_id
  Create a recurring calendar event.
- scale_service(service: str [required], replicas: int [required], region: str [required]) -> dict with new_replica_count and status
  Scale a service to a target number of replicas.
- authenticate_oauth(token_url: str [required], client_id: str [required], client_secret: str [required], scope: str = '') -> dict with access_token and expires_in
  Obtain an OAuth2 access token using client credentials.

In the calendar system the authentication has expired requiring a refresh before any event creation can occur. The relevant endpoint https://api.authprovider.com/refresh must be used with eyJ0eXAiOiJKV1QiLCJhbGciOiJIUzI1NiJ9.eyJzdWIiOiIxMjM0NTY3ODkwIn0 as the refresh token. Following this the recurring event designated as Automated Scaling Window must be created for person_id calendar-bot-5 beginning at 2024-12-01T00:00:00 lasting 15 minutes where the recurrence follows the monthly billing cycle. Once scheduled the scale operation on cache-layer to 12 replicas located in sa-east-1 should be executed. Please determine the ordered sequence of tool calls needed to first address the expired token and then create the associated recurring event incorporating the scaling action as the final step in the workflow.",A Agentic tool use,parameter_construction,a difficult,True,human_professional,planning_agentic,A Agentic tool use
q-corpus-1/q100_6/3,q100_6,3,"In the Toy Story films, why do the toys hold perfectly still whenever a human enters the room?",7 Arts & recreation,"79 Sports, games and entertainment",a very simple,False,agent_structured,world_knowledge,7 Arts & recreation
q-corpus-1/q100_6/4,q100_6,4,"In determining the size of the possible syllable inventory for phonological analysis in this language, there are 8 consonants and 5 vowels available. Syllables follow a CVC template where the two consonants must be different. Calculate the total number of valid syllables under these constraints.",4 Language,40 Language,a simple,False,human_professional,math_reasoning,4 Language
q-corpus-1/q100_6/5,q100_6,5,"I've been reading about the Columbian Exchange and something seems contradictory to me. European populations apparently benefited enormously from New World crops like potatoes and maize, which helped fuel population growth in the 18th century. But the indigenous population of the Americas collapsed by something like 80-90% within a century of contact. I understand diseases like smallpox were the main killer, but I've also read that this collapse wasn't purely accidental -- forced labor systems like the encomienda, and the disruption of native agriculture and social structures, made epidemics far deadlier than they otherwise would have been. Can you explain how the biological exchange (disease, crops) and the Spanish colonial labor and economic systems worked together to produce such different demographic outcomes in the Old World versus the New? I'd like to understand how these factors reinforced each other rather than treating them as separate causes.",9 History,90 History,a,False,human_casual,world_knowledge,9 History
q-corpus-1/q100_6/6,q100_6,6,"I'm bit-banging a debounced button reader on an ATmega328P (Arduino Uno, 16 MHz, no external libraries). It's wired with a pull-up resistor, so the pin reads HIGH when unpressed and LOW when pressed. I want a press to toggle an LED on pin 13, and I only want one toggle per physical press, no matter how long the button is held or how noisy the mechanical contact is.

Here's my current sketch, which double-toggles on some presses and occasionally misses others:

```cpp
const int buttonPin = 2;
const int ledPin = 13;
int ledState = LOW;

void setup() {
  pinMode(buttonPin, INPUT_PULLUP);
  pinMode(ledPin, OUTPUT);
}

void loop() {
  int reading = digitalRead(buttonPin);
  if (reading == LOW) {
    ledState = !ledState;
    digitalWrite(ledPin, ledState);
  }
}
```

Rewrite this loop() (and add any variables/state you need) to implement proper software debouncing with a 50 ms stability window, so the LED toggles exactly once per press regardless of hold duration or contact bounce. Provide the corrected full sketch.",6 Technology,62 Engineering,a simple,True,human_professional,coding,6 Technology
q-corpus-1/q100_6/7,q100_6,7,"1. Source material: the opening 16 bars of the first movement of Brahms's Symphony No. 4 in E minor (Op. 98), transcribed for solo piano reduction.
2. Target idiom: spectral-microtonal realization for retuned piano (19-EDO) and fixed-media electronics.
3. The transformation must preserve Brahms's developing variation logic — every new pitch-class set must derive from the previous through intervallic compression/expansion, inversion, or retrograde — while mapping his diatonic syntax onto 19-EDO's neutral thirds and subminor sevenths.
4. Electronic layer: granular synthesis sourced exclusively from a single recorded strike of the piano's lowest string (A₀), processed through time-variant formant filters whose center frequencies track the evolving harmonic series of the notated piano part in real time.
5. Formal constraint: the 16-bar span must be stretched to exactly 4'33"" without tempo rubato — only through sustained resonance, spectral decay tails, and inserted silences that mirror the rests in Brahms's original phrasing.
6. Notation: produce a two-staff score (piano + electronics trigger stave) using Sagittal microtonal accidentals for 19-EDO, with precise temporal coordinates (seconds.milliseconds) for every electronic event.

Compose the complete 4'33"" realization meeting all six conditions.",7 Arts & recreation,78 Music,a very difficult,False,agent_structured,creative_synthesis,7 Arts & recreation
q-corpus-1/q100_6/8,q100_6,8,"Available tools:
- query_database(sql: str [required], database: str [required], max_rows: int = 100) -> dict with rows and column_names
  Execute a SQL query and return matching rows.
- get_unread_count(user_id: str [required], channel: str = None) -> int
  Get the number of unread messages for a user.
- convert_format(input_path: str [required], output_path: str [required], target_format: str, enum: ['csv','json','xml','yaml','pdf'] [required]) -> str output path
  Convert a file between formats.

Need a small script for a daily report. In the Postgres `msg_archive` there's a table `messages` with user_id, channel, read_status, and timestamp. I want the list of user_ids who have more than 2 unread messages total across all channels in the last day—that means grouping by user_id and counting where read_status = 'unread'. For each of those user_ids I also want to turn their per-user preference file (stored as .xml in /prefs) into a .PDF so I can mail it to the account team. Please write a routine that first runs the SQL query to collect the user_ids with the aggregated unread counts, then for each id performs the XML→PDF conversion. I'll pass the database credentials and the pref directory separately.",A Agentic tool use,tool_selection,a,True,human_casual,planning_agentic,A Agentic tool use
q-corpus-1/q100_6/9,q100_6,9,"I have a new set of high‑precision Sr–Nd–Pb–He isotope measurements from fresh basaltic glasses dredged at 12° N on the East Pacific Rise (see the table below). The claim I'm evaluating is that the elevated ³He/⁴He ratios together with the low ⁸⁷Sr/⁸⁶Sr and high ¹⁴³Nd/¹⁴⁴Nd point to a contribution from a deep, relatively undegassed mantle plume rather than from recycled oceanic crust. Could you give me the strongest arguments for and against this plume‑origin interpretation, using only the data provided?

Sample | ⁸⁷Sr/⁸⁶Sr | ¹⁴³Nd/¹⁴⁴Nd | ²⁰⁶Pb/²⁰⁴Pb | ³He/⁴He (R/Ra)
A | 0.70255 | 0.51310 | 18.6 | 12.5
B | 0.70258 | 0.51308 | 18.7 | 13.1
C | 0.70253 | 0.51312 | 18.5 | 11.8",5 Science,55 Earth sciences and geology,a very difficult,False,human_casual,world_knowledge,5 Science
q-corpus-1/q100_6/10,q100_6,10,"In a fieldwork session on a previously undocumented language (ISO 639-3: xyz), I recorded the following verb forms from a single speaker. The language exhibits templatic morphology with root-and-pattern structure. All forms are cited in broad phonemic transcription; stress is not marked. The verb root is consonantal √k-t-b 'write'. The attested forms:

1. katab        'he wrote'
2. ktibat       'she wrote'
3. yaktub       'he writes'
4. taktub       'she writes'
5. kutub        'books' (plural noun)
6. maktab       'office' (noun)
7. kitabat      'writing' (verbal noun)
8. yaktib       'he dictates'
9. taktib       'she dictates'
10. muktab      'dictated (passive participle)'

The speaker also produced these forms with the root √d-r-s 'study':
11. daras       'he studied'
12. drisat      'she studied'
13. yadrus      'he studies'
14. tadrus      'she studies'
15. durus       'lessons' (plural noun)
16. madras      'school' (noun)
17. dirasat     'studying' (verbal noun)
18. yadris      'he teaches'
19. tadris      'she teaches'
20. mudris      'teacher' (active participle)

Using only the data above, derive the complete set of vocalic melodies and CV-templates for the seven morphological categories represented (perfective, imperfective active, imperfective causative, active participle, passive participle, verbal noun, and noun of place/instrument). State each melody as a sequence of vowel positions (V₁, V₂, V₃) aligned to the three root consonants, and each template as a CV-skeleton with explicit epenthetic vowels. Prove that your analysis accounts for all twenty forms without positing root-allomorphy or morphophonological rules beyond template–melody association and a single, stated syllabification constraint.",4 Language,40 Language,a difficult,True,human_professional,math_reasoning,4 Language
q-corpus-1/q100_6/11,q100_6,11,"Available tools:
- process_payment(amount: float [required], currency: str [required], customer_id: str [required], payment_method: str, enum: ['credit_card','bank_transfer','paypal'] [required], description: str = '') -> dict with transaction_id and status
  Process a payment from a customer.

Known issues:
- Warning: process_payment can timeout after 30 seconds if the underlying service is under heavy load. Plan for retries.

Charge customer CUST-4471 $149.99 USD via credit_card for ""annual plan renewal"". The payment service has been throwing 30s timeouts intermittently all afternoon — if the first call times out, retry the same charge once before reporting. Need a confirmed transaction_id when done.",A Agentic tool use,single_call_error_anticipation,a very simple,True,agent_structured,planning_agentic,A Agentic tool use
q-corpus-1/q100_6/12,q100_6,12,I'm proofreading a manuscript and keep seeing 'between you and I'. My instinct says it's wrong but it sounds so formal. What's the correct pronoun to use there?,4 Language,40 Language,a very simple,False,human_casual,world_knowledge,4 Language
q-corpus-1/q100_6/13,q100_6,13,"Hey, I've been thinking about a track where the lyrics follow a twelve-tone row by cycling through twelve distinct phonetic categories without repeating any until the sequence completes, all while the music uses dissonant harmonies and irregular rhythms from the expressionist era. This is for conveying the disjointed nature of contemporary existence. Could you compose some sample lyrics that adhere to these compositional rules?",7 Arts & recreation,78 Music,a difficult,False,human_casual,creative_synthesis,7 Arts & recreation
q-corpus-1/q100_6/14,q100_6,14,"Our county health department is tracking a respiratory outbreak in a long-term care facility (142 residents). We have daily incident symptom-onset counts for 14 days and we want a rolling estimate of the effective reproduction number R_t using the Cori et al. method: R_t = I_t / sum_{s=1}^{k} I_{t-s} * w_s, where w_s is the discretized serial interval distribution over the previous k days.

The script below is what our epidemiologist intern wrote, but the output is obviously wrong — R_t is negative on two days and above 40 on another, which makes no sense given the data. I can't tell whether the problem is in the serial interval vector, the date alignment, or the rolling window itself. Here is the code verbatim:

```python
import numpy as np
import pandas as pd

onset = pd.Series(
    [0, 1, 0, 2, 3, 5, 8, 9, 11, 7, 4, 2, 1, 0],
    index=pd.date_range(""2024-03-04"", periods=14),
)

# Discretized serial interval (gamma, mean 3.6 days, SD 1.6 days)
si = np.array([0.05, 0.22, 0.33, 0.24, 0.11, 0.05])
k = len(si)

def estimate_rt(onset, si, window=3):
    I = onset.values.astype(float)
    rt = {}
    for t in range(len(I)):
        num = I[max(0, t - window + 1):t + 1].sum()
        denom = sum(
            I[t - s] * si[s] for s in range(1, k + 1) if t - s >= 0
        )
        rt[onset.index[t]] = num / denom if denom > 0 else np.nan
    return pd.Series(rt)

print(estimate_rt(onset, si))
```

Debug this script, explain exactly where it goes wrong relative to the Cori method (there is more than one issue, including how the rolling numerator is constructed and how the serial interval weights are indexed), and rewrite it so it produces a sensible daily R_t series for the 14 days — clearly handling the first few days where the estimate is undefined. Return the corrected code with brief inline comments marking each fix.",6 Technology,61 Medicine and health,a difficult,True,human_casual,coding,6 Technology
q-corpus-1/q100_6/15,q100_6,15,"A comparative fifth-century gallery currently splits the story: collapse for lowland Britain, transformation for northern Gaul. Supply arguments for and against the expert claim that this split is empirically justified—that Roman fiscal, ceramic, and urban systems in lowland Britain failed earlier and more completely than those north of the Loire, so the Late Antique transformation model cannot be extended across the Channel. Constrain the case to: shutdown dates of the Oxfordshire, New Forest, and Alice Holt potteries versus continued Argonne ware and African Red Slip in northern Gaul; coin loss after the Honorius issues and the absence of a British successor precious-metal coinage versus Merovingian minting; pollen and colluvial evidence for woodland regeneration and reduced arable in eastern England versus cereal continuity in the Paris basin and middle Rhine; and intramural sequences at Canterbury, Wroxeter, and Verulamium read against episcopal-city continuity at Tours, Reims, and Trier. Treat Gildas as a source that can be used on either side. Do not restate that Rome fell in 476.",9 History,90 History,a very difficult,False,agent_structured,world_knowledge,9 History
q-corpus-1/q100_6/16,q100_6,16,"Analyze the claim that the specific geography of the Eurasian Steppe—specifically the lack of natural barriers and the presence of fragmented grasslands—was the primary catalyst for the development of the ""Steppe Empire"" cycle of nomadic conquest, rather than the internal socio-political structures of the Xiongnu or Mongol confederations. Provide a detailed set of arguments supporting the environmental determinism of this claim, as well as a set of counter-arguments emphasizing the agency of political centralization and military innovation.",9 History,91 Geography,a difficult,False,agent_structured,world_knowledge,9 History
q-corpus-1/q100_6/17,q100_6,17,"I'm looking over some unpublished magnetics from a recent icebreaker survey of the Gakkel Ridge. They pick the Brunhes-Matuyama boundary (0.78 Ma) at 4.1 km from the spreading axis. If we take the hypotheses that the identification is correct, spreading has been constant in rate, and perfectly symmetric, what's the half-spreading rate in mm/yr? Just the derivation from those numbers.",5 Science,55 Earth sciences and geology,a,False,human_casual,math_reasoning,5 Science
q-corpus-1/q100_6/18,q100_6,18,"Annotators keep collapsing 2 and 3 into a single “bad pronoun” tag. They are different violations; the annotation guidelines have to separate them. Coindexation marks the intended bound-variable reading; a star means that reading is unavailable.
1. Who_i t_i called his_i mother?
2. *Who_i did his_i mother call t_i?
3. *Who_i did he_i say Mary called t_i?
4. Which senator_i did you claim t_i had bribed his_i brother?
5. *Which senator_i did his_i brother claim you had bribed t_i?
Account for the bound-variable judgments in 1–5 by stating the c-command requirement on bound pronouns and the distinction between weak and strong crossover.",4 Language,40 Language,a difficult,False,agent_structured,world_knowledge,4 Language
q-corpus-1/q100_6/19,q100_6,19,"I am preparing a briefing for a municipal housing office that cites a 30% increase in unsheltered homelessness as evidence of policy failure. What official measurement instruments and definitions does HUD use to count unsheltered homelessness in the Point-in-Time count, and how do those methodological constraints affect the validity of year-over-year comparisons for that claim?",3 Social Science,36 Social problems,a simple,False,human_professional,world_knowledge,3 Social Science
q-corpus-1/q100_6/20,q100_6,20,"Available tools:
- get_exchange_rate(from_currency: str [required], to_currency: str [required]) -> dict with rate and timestamp
  Get the current exchange rate between two currencies.
- book_appointment(person_id: str [required], start_time: str [required], duration_min: int [required], title: str [required], attendees: list[str] = []) -> dict with appointment_id
  Book an appointment in a person's calendar.

Known issues:
- Warning: book_appointment can timeout after 30 seconds if the underlying service is under heavy load. Plan for retries.

I need to secure an appointment slot for user ID: 8892.
The meeting details are:
- Title: ""Project Delta Sync""
- Start Time: 2023-10-25T10:00:00Z
- Duration: 45 minutes
- Attendees: []

The booking service is known to timeout after 30 seconds during high load periods. Before executing the request, incorporate a failure strategy to handle this specific risk. If the initial attempt fails, initiate a retry mechanism with an exponential backoff until the expiration of the 30-second window. I require the `appointment_id` from the final successful establishment. Do not stop to report the timeout.",A Agentic tool use,single_call_error_anticipation,a simple,True,human_professional,planning_agentic,A Agentic tool use
q-corpus-1/q100_6/21,q100_6,21,"My sock drawer has 10 black socks and 5 white socks all jumbled together. If I reach in without looking and pull out one sock at random, what is the probability it is white?",5 Science,51 Mathematics,a very simple,False,human_casual,math_reasoning,5 Science
q-corpus-1/q100_6/22,q100_6,22,"Available tools:
- get_order_details(order_id: str [required]) -> dict with order details
  Get full details of an order including line items.
- apply_discount(order_id: str [required], discount_code: str [required]) -> dict with new_total and discount_amount
  Apply a discount code to an order.
- http_delete(url: str [required], headers: dict = {}) -> dict with status_code
  Make an HTTP DELETE request.
- send_template_email(to: list[str] [required], template_name: str [required], variables: dict [required]) -> dict with message_id
  Send an email using a named template with variable substitution.

Please resolve the customer’s delayed-shipment case for the order referenced by warehouse ticket `ORD-48271`. First retrieve the complete order details, then apply the approved discount code `RECOVERY15`. After the discount succeeds, email the customer at the address returned in the order details using the `order_update` template. Include the order identifier, original total, discount amount, new total, and a concise summary of the ordered items in the template variables; do not send the email if the discount cannot be applied.",A Agentic tool use,tool_selection,a difficult,True,human_professional,planning_agentic,A Agentic tool use
q-corpus-1/q100_6/23,q100_6,23,"I'm trying to work out a quantitative structure-function relationship for the loop of Henle and I keep getting tangled. Here's the simplified steady-state model I want to use for a derivation:

- The descending limb is permeable to water, impermeable to NaCl.
- The ascending limb is impermeable to water but actively extrudes NaCl into the surrounding interstitium.
- Active transport creates a fixed 200 mOsm/L osmotic gap between the ascending-limb tubular fluid and the local interstitium at every depth.
- The descending-limb tubular fluid and the local interstitium are in osmotic equilibrium at every depth (water moves freely to balance).
- Tubule radius r = 15 μm, single-nephron flow rate Q = 10 nL/min, fluid enters the descending limb at the cortico-medullary junction at 300 mOsm/L.

From these hypotheses, derive the osmolality profile along the loop as a function of depth x. Then tell me what total loop length L is required to reach a papillary interstitial osmolality of 1200 mOsm/L. I want to see how the active-transport ""single effect"" (200 mOsm/L step) and the geometry (length, flow, radius) jointly set the maximum concentrating ability of the system.",5 Science,57 Biology,a,False,human_casual,math_reasoning,5 Science
q-corpus-1/q100_6/24,q100_6,24,"I'm the head of technical services at a mid-sized public library, and I need a definitive answer to a dispute with our director before the accreditation review next month. Our local-history workflow requires two cataloguers, R. and M., to independently decide for every incoming item whether it receives the ""Local History"" subject heading. The review measures their consistency with Cohen's κ against the Landis–Koch scale, and the auditor requires ""substantial"" agreement, κ ≥ 0.61, on this binary decision.

The director's position is that our current low score is just an alignment problem fixable by retraining: she claims that as long as each cataloguer keeps their overall assignment rate stable — which our collection-development policy effectively locks in, since those rates follow from the acquisition profile — the item-level decisions can be re-aligned to push κ to 0.61 or higher. She put it as ""the marginals don't cap how high κ can go."" I think she's wrong, but I can't prove it, and the retraining vendor is quoting us a price based on her assumption.

From last quarter's dual-review sample:

1. The sample contains 400 records, each independently judged by both cataloguers.
2. R. assigned the heading to 80 of the 400 records (marginal rate 0.20, fixed by policy for the coming year).
3. M. assigned the heading to 200 of the 400 records (marginal rate 0.50, fixed by policy for the coming year).
4. The two cataloguers both assigned the heading to the same 60 records this quarter; on the current data their κ works out to 0.20.
5. κ is the standard chance-corrected coefficient, κ = (P_o − P_e)/(1 − P_e), where P_o is the observed agreement rate and P_e is the agreement rate expected from the marginals under statistical independence of the two cataloguers.
6. The director's claim to be evaluated: by retraining alone, the joint decisions can be rearranged so that κ ≥ 0.61, while R.'s marginal rate stays exactly 0.20 and M.'s stays exactly 0.50 across the 400 records.

Disprove or confirm the director's claim by computing the exact maximum value of κ attainable over all joint distributions of the two cataloguers' decisions consistent with the fixed marginals in 2 and 3, proving that your value is the true maximum (i.e., that no admissible rearrangement exceeds it and at least one achieves it), and stating explicitly whether 0.61 is reachable.","0 Computer science, information, and general works",02 Library and information sciences,a very difficult,False,human_casual,math_reasoning,"0 Computer science, information, and general works"
q-corpus-1/q100_6/25,q100_6,25,"I’m debugging a small clinical decision-support script for a primary-care dashboard. A patient vignette: serum calcium 7.9 mg/dL, albumin 2.1 g/dL. Our function should return corrected calcium using the common formula: corrected calcium = measured calcium + 0.8 * (4.0 - albumin). The UI says 9.4 mg/dL, but the code returns 7.6. Here is the Python:

def corrected_calcium(ca_mg_dl, albumin_g_dl):
    return ca_mg_dl + 0.8 * (albumin_g_dl - 4.0)

print(corrected_calcium(7.9, 2.1))

Can you debug this and explain the basic idea behind why the correction term is added when albumin is low? Please give the corrected function and the value it should return for these labs.",6 Technology,61 Medicine and health,a simple,False,human_professional,coding,6 Technology
q-corpus-1/q100_6/26,q100_6,26,"Let $S_n$ denote the symmetric group on $[n] = \{1, 2, \dots, n\}$. A permutation $\pi \in S_n$ is called parity-preserving if $\pi(i) \equiv i \pmod 2$ for all $i \in [n]$, and $k$-bounded if $|\pi(i) - i| < k$ for all $i \in [n]$. Let $P(n, k)$ denote the set of all parity-preserving, $k$-bounded permutations in $S_n$.

For an even universe size $n = 2m$ ($m \ge 1$):

1. Prove that $|P(2m, 3)| = F_{m+1}^2$, where $F_1 = 1, F_2 = 1, F_3 = 2, \dots$ are the Fibonacci numbers.
2. Evaluate $|P(2m, 4)|$ and determine whether the equality $|P(2m, 2r)| = |P(2m, 2r-1)|$ holds for all positive integers $r \ge 1$.
3. For $k = 5$, formulate the bijection between $P(2m, 5)$ and pairs of permutations in $S_m$ with bandwidth at most 2, and determine the closed rational form of the generating function $\sum_{m \ge 0} \sqrt{|P(2m, 5)|}\, x^m$.",5 Science,51 Mathematics,a difficult,False,agent_structured,math_reasoning,5 Science
q-corpus-1/q100_6/27,q100_6,27,"Available tools:
- retrieve_document(doc_id: str [required], collection: str [required]) -> dict with full document content
  Retrieve a full document by its ID.
- get_customer(customer_id: str [required], include_orders: bool = False, include_notes: bool = False) -> dict with customer details
  Retrieve a customer profile by ID.

Our incident triage stream just flagged a billing inconsistency tied to customer `C-4471`. Per data-minimization rules we must surface the account profile with its order history but without internal agent notes. Select the tool that retrieves a customer profile by identifier and explain in prose how its arguments are derived from these facts: the customer identifier is `C-4471`, the order inclusion flag should be set true, and the notes inclusion flag false. The outcome is a concise explanation that names this tool and states those parameter values, without emitting a fully built invocation.",A Agentic tool use,parameter_construction,a simple,True,human_professional,planning_agentic,A Agentic tool use
q-corpus-1/q100_6/28,q100_6,28,"I have a patient whose acid-base picture I can't fully untangle, and I want a rigorous, calculation-backed diagnosis rather than a gestalt read, because the treatment decisions (bicarbonate? naloxone? insulin alone?) hinge on which disorders are actually present.

The patient is a 54-year-old with type 1 diabetes and chronic low back pain managed with sustained-release oxycodone. She was brought in somnolent after three days of vomiting and polyuria. She takes hydrochlorothiazide for edema. On exam she is obtunded but breathing slowly and shallowly at 8 breaths/min.

Arterial blood gas: pH 7.21, PaCO2 34 mmHg, HCO3 13 mEq/L.
Serum chemistry: Na 140, K 3.1, Cl 100, HCO3 13 mEq/L (both measured on the chemistry panel, so assume the two bicarbonates are concordant), glucose 520 mg/dL, BUN 34 mg/dL, creatinine 1.8 mg/dL, albumin 2.4 g/dL, lactate 3.1 mEq/L. Urine ketones are large-positive. Measured serum osmolality is 322 mOsm/kg. Serum ethanol and salicylate are zero. She has no history of lithium, methanol, ethylene glycol, or toluene exposure, and no Evidence of sepsis.

I need one deliverable: the complete acid-base diagnosis, with every component proven or disproven by explicit quantitative argument, using only the data above. Specifically, work through and show the numbers for:

1. Internal consistency of the blood gas itself — verify that the reported pH, PaCO2, and HCO3 can coexist (Henderson–Hasselbalch with pK 6.1 and CO2 solubility 0.03), so we know the sample isn't corrupted before we reason further.
2. The albumin-corrected anion gap, including by how much hypoalbuminemia distorts the raw gap, and whether the corrected gap supports a high-anion-gap process once you also account for the lactate and the expected ketoacid contribution.
3. Whether a toxic alcohol or other unmeasured osmole is hiding behind the ketoacidosis — compute the calculated and measured osmolality and the osmolar gap, and state whether the gap excludes or merely fails to confirm this.
4. The delta–delta (delta ratio) analysis, given the albumin-corrected gap, to determine whether there is a second, simultaneous metabolic disturbance — and identify which one, tying it to at least one feature of her history that would explain it mechanistically.
5. Whether the respiratory compensation is appropriate: compute the expected PaCO2 from the appropriate compensation equation for her bicarbonate, compare it against the measured PaCO2 with a ±2 mmHg tolerance, and determine whether a separate respiratory disorder exists despite the fact that the pH is acidemic and ""compensation looks incomplete"" at first glance.
6. Whether the potassium value of 3.1 is consistent with the acid-base status or implies something additional about her total body potassium given the diuretic and vomiting.

Conclude with a single ranked list of every disorder present (primary and compensatory processes), each labeled as proven or excluded by your calculations, and note anywhere the data are internally ambiguous so I know what to recheck. Show every intermediate number — I want to audit the arithmetic against her chart.",6 Technology,61 Medicine and health,a very difficult,False,human_casual,math_reasoning,6 Technology
q-corpus-1/q100_6/29,q100_6,29,"```python
class IntTokenBucket:
    def __init__(self, capacity_tokens, rate_per_sec):
        self.capacity = capacity_tokens
        self.rate = rate_per_sec
        self.tokens = capacity_tokens
        self.last_ns = 0

    def take(self, n, now_ns):
        # capacity_tokens, rate_per_sec, n: nonnegative ints.
        # now_ns: nonnegative, nondecreasing.
        delta = now_ns - self.last_ns
        whole = (delta * self.rate) // 1_000_000_000
        if whole:
            self.tokens = min(self.capacity, self.tokens + whole)
            self.last_ns = now_ns
        if self.tokens >= n:
            self.tokens -= n
            return True
        return False
```

At `rate_per_sec = 3`, successive `take(1, ...)` calls at `0`, `333_333_333`, `666_666_666`, and `999_999_999` grant only the first call, although the intended refill is 3 tokens per second and the final call should have enough balance to grant.

Repair the class so it honors the stated integer-rate refill without losing fractional refill time, while never exceeding capacity and never granting a token that has not accrued. Use integer arithmetic only. Provide one complete corrected Python class with the same public method names. Include exactly one comment line at the top of the class, beginning `COUNTERFACTUAL:`, stating what would follow if `now_ns` were allowed to decrease between calls under your implementation.","0 Computer science, information, and general works","00 Computer science, knowledge, and systems",a difficult,True,agent_structured,coding,"0 Computer science, information, and general works"
q-corpus-1/q100_6/30,q100_6,30,"Recent phylogenomic and morphological reassessments have revived the hypothesis that Ctenophora, rather than Porifera, represents the sister lineage to all other extant Metazoa. If true, this claim entails that the last common metazoan ancestor possessed neuromuscular and developmental toolkit characters that were either secondarily lost in Porifera or convergently acquired in Ctenophora relative to Cnidaria and Bilateria. Evaluate the claim using the evidentiary landscape below:

| Line of Evidence | Pro-Ctenophora-Sister Signal | Major Caveat or Alternative Interpretation |
|---|---|---|
| Concatenated nuclear supermatrices | Recovered Ctenophora-sister with high bootstrap support in several large datasets (Ryan et al., 2013; Whelan et al., 2017) | Long-branch attraction between Ctenophora and outgroups; compositional heterogeneity; sensitivity to taxon and gene sampling |
| Rare genomic changes (intron positions, domain architectures) | Shared derived losses suggesting a clade of Porifera + non-ctenophore metazoans | Sparse sampling in early-branching lineages; homoplasy in intron gain/loss |
| Mitochondrial genome architecture | Highly derived ctenophore mtDNA lacking obvious poriferan similarities | Extreme derivedness may erase ancestral signals; Porifera itself exhibits extensive mtDNA variability |
| Morphological and developmental characters | Ctenophora possesses true muscle cells, a synaptic nerve net, and an apical sense organ absent in Porifera | Convergent evolution of neuromuscular traits remains possible; sponge simplicity may be secondary rather than plesiomorphic |

Provide a critical evaluation structured as follows:

1. Construct two independent arguments *in favor of* the claim that Ctenophora is the sister group to all other Metazoa. Each argument must integrate phylogenomic logic with one or more rows from the table above.

2. Construct two independent arguments *against* that same claim. One argument must center on systematic biases in molecular phylogenetics (e.g., site-heterogeneous substitution models, coalescent vs. concatenation behavior, long-branch attraction). The second argument must center on morphological homology and the evolutionary interpretation of the metazoan ground plan (addressing, for instance, the homology of cnidocytes and colloblasts, or the developmental origins of the ctenophore aboral organ).

3. Conclude with a concise synthesis stating which position currently withstands scrutiny better, explicitly justifying your verdict by weighing model fit against biological plausibility rather than by tallying published studies.

Your analysis should demonstrate expert familiarity with both invertebrate anatomy and advanced phylogenetic inference.",5 Science,59 Animals (Zoology),a very difficult,False,human_professional,world_knowledge,5 Science
q-corpus-1/q100_6/31,q100_6,31,"Can you give me a brief, plain-language explanation of how the burden of a payroll tax actually gets split between workers and employers when the labor market isn't perfectly competitive? I'm looking at nurses in a tight hospital market where employers have a lot of bargaining power, and the textbook 50/50 split doesn't seem to match what I'm seeing. Why does statutory incidence matter less than the relative elasticities here?",3 Social Science,33 Economics,a simple,False,human_casual,world_knowledge,3 Social Science
q-corpus-1/q100_6/32,q100_6,32,"I have a producer-consumer setup using a bounded buffer protected by a mutex and two condition variables (not_full and not_empty) in POSIX threads. Under load with multiple producers and consumers, the program occasionally deadlocks. Here's the relevant code:

```c
pthread_mutex_t mutex = PTHREAD_MUTEX_INITIALIZER;
pthread_cond_t not_full = PTHREAD_COND_INITIALIZER;
pthread_cond_t not_empty = PTHREAD_COND_INITIALIZER;
int count = 0, in = 0, out = 0;
int buffer[BUFFER_SIZE];

void* producer(void* arg) {
    while (1) {
        int item = produce_item();
        pthread_mutex_lock(&mutex);
        if (count == BUFFER_SIZE)
            pthread_cond_wait(&not_full, &mutex);
        buffer[in] = item;
        in = (in + 1) % BUFFER_SIZE;
        count++;
        pthread_cond_signal(&not_empty);
        pthread_mutex_unlock(&mutex);
    }
}

void* consumer(void* arg) {
    while (1) {
        pthread_mutex_lock(&mutex);
        if (count == 0)
            pthread_cond_wait(&not_empty, &mutex);
        int item = buffer[out];
        out = (out + 1) % BUFFER_SIZE;
        count--;
        pthread_cond_signal(&not_full);
        pthread_mutex_unlock(&mutex);
        consume_item(item);
    }
}
```

Identify exactly why this deadlocks/misbehaves with multiple producers and consumers, and rewrite the producer and consumer functions so the buffer works correctly under concurrent access.","0 Computer science, information, and general works","00 Computer science, knowledge, and systems",a simple,False,human_professional,coding,"0 Computer science, information, and general works"
q-corpus-1/q100_6/33,q100_6,33,Just sat down and counted 18 beats in 15 seconds. What’s my heart rate per minute?,6 Technology,61 Medicine and health,a very simple,False,human_casual,math_reasoning,6 Technology
q-corpus-1/q100_6/34,q100_6,34,"A café sign reads: “Kids make nutritious snacks.”

Can you explain the sentence’s syntactic ambiguity and its two possible meanings?",4 Language,40 Language,a very simple,True,human_casual,world_knowledge,4 Language
q-corpus-1/q100_6/35,q100_6,35,"Hey, can you draft a short letter in first person from a Cherokee farmer in north Georgia in late 1838 to a cousin who stayed behind, showing how the Indian Removal Act enforcement and the order to march to Oklahoma is affecting his family's decisions about food, shelter, and staying together? Keep it in his voice, mention the stockade at Ross's Landing and the winter ahead, and reference the rations being withheld.",9 History,90 History,a,False,human_casual,creative_synthesis,9 History
q-corpus-1/q100_6/36,q100_6,36,"I am trying to reconcile an unexpectedly slower, more exergonic outer-sphere electron-transfer reaction with Marcus theory.

1. Two one-electron outer-sphere electron-transfer reactions occur at 298.15 K in the same solvent and at identical donor–acceptor separation, with 1 M standard states and negligible work-term corrections.
2. For reaction I, the standard reduction potentials are \(E^\circ(\mathrm{A^+/A})=+0.40\ \mathrm{V}\) and \(E^\circ(\mathrm{D^+/D})=-0.40\ \mathrm{V}\).
3. For reaction II, the corresponding potentials are \(E^\circ(\mathrm{A^+/A})=+1.40\ \mathrm{V}\) and \(E^\circ(\mathrm{D^+/D})=-1.40\ \mathrm{V}\).
4. The total classical reorganization energy is \(\lambda=1.00\ \mathrm{eV}\) for both reactions, and the electronic coupling and nuclear-frequency prefactor are identical.
5. Use \(\Delta G^\circ=-F\Delta E^\circ\), \(1\ \mathrm{eV\ molecule^{-1}}=96.485\ \mathrm{kJ\,mol^{-1}}\), \(k_\mathrm{B}T=0.0256926\ \mathrm{eV}\), and the nonadiabatic Marcus relations
\[
\Delta G^\ddagger=\frac{(\lambda+\Delta G^\circ)^2}{4\lambda},
\qquad
k\propto\exp\!\left(-\frac{\Delta G^\ddagger}{k_\mathrm{B}T}\right).
\]

Compute the rate constant ratio \(k_{\mathrm{I}}/k_{\mathrm{II}}\).",5 Science,54 Chemistry,a very difficult,False,human_casual,math_reasoning,5 Science
q-corpus-1/q100_6/37,q100_6,37,"The central bank is targeting a real interest rate of 1.5% to cool down an overheating housing market, and households currently expect inflation to be 3%. Using the Fisher equation, calculate the nominal interest rate the central bank should set to achieve this target.",3 Social Science,33 Economics,a very simple,False,human_casual,math_reasoning,3 Social Science
q-corpus-1/q100_6/38,q100_6,38,"Available tools:
- assign_ticket(ticket_id: str [required], agent_id: str [required]) -> bool
  Assign a support ticket to an agent.
- parse_json_response(response_body: dict [required], json_path: str [required]) -> extracted value
  Extract a value from a nested JSON response using a dot-path.
- update_ticket_status(ticket_id: str [required], status: str, enum: ['open','in_progress','waiting','resolved','closed'] [required], comment: str = '') -> bool
  Update the status of a support ticket.
- create_notification(user_id: str [required], title: str [required], body: str [required], action_url: str = '') -> dict with notification_id
  Create an in-app push notification for a user.

Known issues:
- Note: parse_json_response may return a rate-limit error (HTTP 429) if called more than 3 times per minute.

We've already burned two parse_json_response calls this minute on the morning queue, so the next one is the last before a 429. I cannot wait a cooldown — this chargeback window closes in a few minutes. From the webhook body below, take the single remaining call and pull the nested primary email at data.result.customer.contact.primary_email. Do not probe sibling paths.

{""event"":""dispute.opened"",""data"":{""result"":{""ticket_ref"":""T-88421"",""customer"":{""id"":""cus_9k2"",""contact"":{""primary_email"":""lea.voss@example.org"",""alt_email"":""lvoss.billing@example.org""},""account"":{""region"":""eu-central""}},""flags"":{""vip"":false}}}}",A Agentic tool use,single_call_error_anticipation,a difficult,True,human_casual,planning_agentic,A Agentic tool use
q-corpus-1/q100_6/39,q100_6,39,"I was tracing how the English phrase ""going to"" shifted from indicating physical movement to marking future tense, often contracting into ""gonna"". A text I was reading labeled this historical shift as ""grammaticalization"". What is grammaticalization?",4 Language,40 Language,a very simple,False,agent_structured,world_knowledge,4 Language
q-corpus-1/q100_6/40,q100_6,40,"Available tools:
- paginate_results(url: str [required], headers: dict = {}, page_param: str = 'page', per_page: int = 50, max_pages: int = 10) -> list of all result items
  Fetch all pages from a paginated API endpoint.
- refresh_auth_token(token_url: str [required], refresh_token: str [required]) -> dict with access_token and expires_in
  Refresh an expired authentication token.
- web_search(query: str [required], num_results: int = 10, region: str = 'global') -> list of result dicts with title, url, snippet
  Search the web and return top results.

Hey, I'm putting together some data for the analysis and need every item from the paginated endpoint https://api.service.com/records. The token expired on the last try, so I need to use the refresh at https://auth.service.com/token with the refresh token 'eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9'. How would you get the complete results?",A Agentic tool use,tool_selection,a,True,human_casual,planning_agentic,A Agentic tool use
q-corpus-1/q100_6/41,q100_6,41,"City of Riverview proposal - Ordinance 14-B: ""No landlord may charge more than $1,200 per month for a 1-bedroom apartment.""

Housing Department data for 1-bedroom units:
- Current equilibrium rent: $1,500 with 10,000 units rented.
- At $1,200: quantity demanded is 13,000 households, quantity supplied is 7,000 units.

Using the supply and demand model's concept of a binding price ceiling, what does the model predict will happen in this rental market and why wouldn't the lower legal price clear the market?",3 Social Science,33 Economics,a simple,True,human_casual,world_knowledge,3 Social Science
q-corpus-1/q100_6/42,q100_6,42,"I’m stuck deciding how to handle the team that isn’t replying to morning invites. The VP of Ops recommends a year-end bonus to restore motivation, while the psychologist consultant insists we stop measuring output and instead give them full control over their schedules. I need a clear explanation of the gap between Maslow’s Hierarchy of Needs and Self-Determination Theory here. Specifically, does addressing safety and finances with the bonus fix the silence as effectively as offering autonomy and choice? I want to know which framework suggests physical rewards create independence versus which says they just mask the lack of competence validation. Please map out the exact psychological prediction for each model if we choose money over workflow freedom.",1 Philosophy and psychology,15 Psychology,a,False,human_casual,world_knowledge,1 Philosophy and psychology
q-corpus-1/q100_6/43,q100_6,43,"Quick sanity check before I send this to the team: we're planning to raise the price of our $10 widget by $1 and we're telling management ""customers pay the whole increase."" We sell 1,000 units/week at $10, and our last price bump from $9 to $10 dropped weekly sales from 1,100 to 1,000 units.

Using that data, can you work out the price elasticity of demand at the $10 price point, then show mathematically what share of a $1 tax-like increase would actually fall on consumers vs. what we'd effectively absorb through lost volume? I just need the elasticity number and the consumer/producer incidence split laid out step by step so I can check my own math matches.",3 Social Science,33 Economics,a very simple,False,human_professional,math_reasoning,3 Social Science
q-corpus-1/q100_6/44,q100_6,44,"Colleagues, the research meeting Thursday needs a one-page internal note on the identification in the municipal-fiber paper from last week. The authors claim that the staggered award of publicly funded gigabit grants between 2014 and 2019 caused a 9–14 percent rise in employer establishment births, operating through a reduction in suppliers’ search and matching frictions rather than through household demand or amenity effects. They show event-study coefficients that are flat pre-treatment after county and year fixed effects plus a vector of pre-period covariates. I need the strongest arguments for treating those grant dates as conditionally exogenous to local entrepreneurial demand, and the strongest arguments against, focused on whether the post-treatment co-movement is the claimed search-cost channel or leftover correlation with unobserved local productivity shocks or simultaneous complementary public investment. Produce a single memo that lays out the points on each side, drawing on the usual staggered-DiD threats (treatment-effect heterogeneity across cohorts, the influence of never-treated units, possible Ashenfelter-style dips in the application process, firm anticipation, and concurrent road or power upgrades) together with any established facts about the timing of municipal broadband relative to local labor-market conditions. Stay strictly on identification and the specific mechanism.",3 Social Science,33 Economics,a very difficult,False,human_professional,world_knowledge,3 Social Science
q-corpus-1/q100_6/45,q100_6,45,"Rewrite the Python function below so it correctly implements a metformin renal-safety rule, and include executable assertions for both the actual case and the counterfactual.

Use the 2021 CKD-EPI creatinine equation:

`eGFR = 142 * min(Scr/k,1)^a * max(Scr/k,1)^-1.200 * 0.9938^age * (1.012 if female else 1)`

where `k=0.7, a=-0.241` for female patients and `k=0.9, a=-0.302` for male patients. Use the latest creatinine for eGFR. Detect AKI when creatinine rises by at least 0.3 mg/dL from any prior result within 48 hours, or reaches at least 1.5 times the lowest prior result within 7 days. Compare unrounded values and treat threshold equality as positive.

If AKI is present, recommend `""hold metformin""` regardless of eGFR. Otherwise return `""contraindicated""` below 30, `""limit to 1000 mg/day and review""` from 30 through 44.999 for an existing prescription, and `""continue""` at 45 or above. Return eGFR rounded to one decimal, the AKI Boolean, and the recommendation. Inputs may be chronologically unordered ISO-8601 timestamps.

```python
def metformin_advice(age, sex, labs, taking_metformin):
    scr = labs[0][1]
    k = 0.9 if sex == ""female"" else 0.7
    a = -0.302 if sex == ""female"" else -0.241
    egfr = 142 * min(scr/k, 1)**a * max(scr/k, 1)**-1.2 * 0.9938**age
    aki = scr - labs[-1][1] > 0.3
    if egfr < 30:
        advice = ""contraindicated""
    elif aki:
        advice = ""hold metformin""
    else:
        advice = ""continue""
    return round(egfr, 1), aki, advice
```

The patient is a 68-year-old woman already taking metformin, with creatinine values `1.0` at `2025-04-01T08:00:00+00:00`, `1.1` at `2025-04-06T20:00:00+00:00`, and `1.4` at `2025-04-08T08:00:00+00:00`. The corrected code should recognize the exactly 0.3 mg/dL rise within 36 hours and hold metformin. If the latest value were instead `1.15`, it should find no AKI and recommend continuing because eGFR is at least 45.",6 Technology,61 Medicine and health,a difficult,False,human_casual,coding,6 Technology
q-corpus-1/q100_6/46,q100_6,46,"Available tools:
- verify_payment(transaction_id: str [required]) -> dict with status, amount, timestamp
  Verify the status of a payment by transaction ID.

A vendor’s emailed receipt may be forged. Call `verify_payment` once with transaction ID `txn_8F3K2L91`, and report the returned status, amount, and timestamp without assuming the payment succeeded.",A Agentic tool use,tool_selection,a very simple,True,agent_structured,planning_agentic,A Agentic tool use
q-corpus-1/q100_6/47,q100_6,47,"Available tools:
- create_notification(user_id: str [required], title: str [required], body: str [required], action_url: str = '') -> dict with notification_id
  Create an in-app push notification for a user.
- semantic_search(query: str [required], collection: str [required], top_k: int = 10) -> list of result dicts with id, text, similarity_score
  Perform semantic vector search on a document collection.
- cancel_appointment(appointment_id: str [required], notify_attendees: bool = True, reason: str = '') -> bool
  Cancel an existing appointment.

Known issues:
- Warning: cancel_appointment can timeout after 30 seconds if the underlying service is under heavy load. Plan for retries.

Kill the Friday 2pm hold — id is appt_9f3c12. Don't ping attendees; the attending already told them in rounds. Reason: OR overrun, case moved to Friday. Booking has been under heavy load and cancel calls hang past 30 seconds, so retry if it times out instead of treating silence as success.",A Agentic tool use,single_call_error_anticipation,a,True,human_casual,planning_agentic,A Agentic tool use
q-corpus-1/q100_6/48,q100_6,48,"I’ve got this heavy-looking stone I picked up on a hike. It has a mass of 960 g, and when I put it in a water-filled graduated cylinder, the level rose from 80 mL to 112 mL. What is the stone’s density in g/mL? Use the density formula and provide the answer with units.",5 Science,50 Science,a simple,False,human_casual,math_reasoning,5 Science
q-corpus-1/q100_6/49,q100_6,49,"I'm finalizing the exhibit text for the Whitman County Pioneers display, and the new research on the McAllister sisters has upended the draft I wrote last month. We've confirmed through land records and a recovered 1889 correspondence bundle that Eliza McAllister wasn't a widow managing her late husband's claim — she was the eldest of four sisters who pooled their inheritance to file adjacent homesteads under the Preemption Act, using her name as the sole filer because the others were under twenty-one. The effect ripples through everything: her 1892 school board candidacy, the ""widow's pension"" application she never actually filed, even the tone of her letters to the territorial governor about irrigation rights. How does this collective filings strategy reshape the narrative of her public advocacy — specifically, rewrite the 150-word exhibit panel to show her actions as coordinated family policy rather than individual grit, without erasing the legal constraints that made the ruse necessary?",9 History,92 Biography and genealogy,a,False,human_professional,creative_synthesis,9 History
q-corpus-1/q100_6/50,q100_6,50,"Hi, I’m finalizing a sketch grammar for a Northwest Caucasian-style language and my reviewer wants a concrete estimate of the size of the lexical space for roots.

The phoneme inventory I am working with is:
Vowels /i e a o u/ with [back] = [-back] for i, e and [+back] for a, o, u.
Consonants 16: p t k q b d g m n ŋ f s ʃ x l r w. Labials are p b m f w.

Phonotactics and root conditions I am assuming:
- Syllable template is CV and a three-syllable root is C1 V1 C2 V2 C3 V3.
- Onset is obligatory, no codas.
- Vowel harmony: V1, V2 and V3 must all have the same [back] value, i.e. all front or all back.
- OCP on vowels: no two adjacent vowels are identical, so V1≠V2 and V2≠V3.
- Consonant distinctness: C1, C2 and C3 are all distinct.
- Labial requirement: the root must contain at least one labial consonant among C1, C2, C3.

Can you give me the exact number of distinct three-syllable roots that satisfy all of these conditions?",4 Language,40 Language,a very difficult,False,human_casual,math_reasoning,4 Language
q-corpus-1/q100_6/51,q100_6,51,"Our market-data feed handler streams about 20,000 small (~40 byte) updates per second to subscribers over a colocated network where one-way delay is ~50µs. We've enabled TCP_NODELAY because the docs warn that Nagle's algorithm will ""add 40ms of latency"" to small writes. But internal tracing shows that with TCP_NODELAY on, p50 round-trip from publish-to-ack is ~90µs while p99 jumps to 3–5ms during bursts, and we're chasing that tail.

How does disabling TCP_NODELAY (letting Nagle coalesce small writes) actually affect the latency distribution of this publish-to-ack path under bursty load, and can it paradoxically reduce p99 in our setup?

Write a small C++ or Rust program that:
- spins up a TCP server and client over loopback with a configurable simulated one-way delay (e.g., 50µs),
- the client bursts N small writes followed by a read-for-ack at a configurable rate,
- measures per-write RTT (write ts → ack ts) for ~100k samples,
- reports p50/p95/p99/p999 with TCP_NODELAY=1 and TCP_NODELAY=0, across burst sizes 1, 8, 64, 256 outstanding writes.

Show me the code and the actual numbers from your run, and explain what you saw — I want measured evidence of whether Nagle is hurting or helping our p99.","0 Computer science, information, and general works","00 Computer science, knowledge, and systems",a,False,human_casual,coding,"0 Computer science, information, and general works"
q-corpus-1/q100_6/52,q100_6,52,"> Original draft for product listing — factual but unstyled:
> ""The chair is constructed from solid oak. It has a dark walnut finish. It supports up to 300 pounds. It is priced at $450. It is built to last for decades.""

*Note: All factual details must be preserved in any revision.*

Please rewrite the draft above for a luxury furniture catalog using **anaphora** as the central rhetorical device to create a persuasive, rhythmic voice, and accompany your revision with a brief explanation of how the anaphoric repetition alters the affective and persuasive appeal for the prospective buyer.",8 Literature,"80 Literature, rhetoric and criticism",a simple,True,human_professional,creative_synthesis,8 Literature
q-corpus-1/q100_6/53,q100_6,53,"I have a pandas cohort pipeline (Python 3.11, pandas 2.x) that computes estimated GFR for ~40k nephrology outpatient records and flags CKD stage. It currently only implements CKD-EPI 2009 (race-based) creatinine equation, but our IRB now requires side-by-side output against CKD-EPI 2021 (race-free) so a reviewer can see how many patients shift stage. I need you to fix the existing function and add the second equation, then produce a comparison.

Input dataframe `df` columns: `patient_id`, `age` (years, float), `sex` (`""M""`/`""F""`), `race` (`""Black""`/`""Other""`), `scr` (serum creatinine, may be in mg/dL or µmol/L — there's a `units` column with values `""mg/dL""` or `""umol/L""` that must be normalized to mg/dL before any calculation, 1 mg/dL = 88.4 µmol/L).

Current (buggy) code:

```python
def ckd_epi_2009(age, sex, race, scr):
    kappa = 0.7 if sex == ""F"" else 0.9
    alpha = -0.329 if sex == ""F"" else -0.411
    min_term = min(scr / kappa, 1) ** alpha
    max_term = max(scr / kappa, 1) ** -1.209
    egfr = 141 * min_term * max_term * (0.993 ** age)
    if sex == ""F"":
        egfr *= 1.018
    if race == ""Black"":
        egfr *= 1.159
    return egfr
```

Bugs I suspect but haven't isolated: unit conversion isn't applied anywhere upstream, and I think the exponent handling for min/max terms might be wrong for edge cases where scr/kappa == 1. Fix this function, add `ckd_epi_2021(age, sex, scr)` (same structure, race term removed, coefficients: kappa/alpha unchanged, multiplier 142, age term 0.9938^age, female multiplier 1.012), and write a vectorized pipeline step that: (1) converts units, (2) computes both eGFR values per row, (3) assigns CKD stage (G1 ≥90, G2 60–89, G3a 45–59, G3b 30–44, G4 15–29, G5 <15) for each equation, (4) outputs a summary table of patient counts whose stage differs between the two equations, broken out by race. Exclude age < 18 from staging (set NaN) but keep raw eGFR values. Show me the corrected code and the summary logic.",6 Technology,61 Medicine and health,a,False,agent_structured,coding,6 Technology
q-corpus-1/q100_6/54,q100_6,54,"Compare the Nile Delta and the Indus Valley on a historical map: how did each river’s flood pattern shape the location and connectivity of early urban settlements? Identify one similarity, one difference, and explain the geographic cause of each.",9 History,91 Geography,a,False,agent_structured,world_knowledge,9 History
q-corpus-1/q100_6/55,q100_6,55,"Excerpt from Samuel Taylor Coleridge's ""The Rime of the Ancient Mariner"":
""The Wedding-Guest sat on a stone:
He stared with open eyes
And listened like a thing for whose
Something to be done was wished.""

Throughout the poem, the Mariner compels the Wedding-Guest to listen to his tale of spiritual alienation and cosmic guilt. The Guest’s initial resistance transforms into a state of hypnotic fixation, while the Mariner’s narrative functions as a ritual of purgation, where the act of storytelling is the only means of achieving momentary relief from his burden.

Construct a set of arguments for and against the claim that the Wedding-Guest serves as a passive surrogate for the reader, rather than a developed character within the narrative arc. Your analysis must address how the Guest's psychological transition from annoyance to ""staring"" affects the poem's rhetorical impact.",8 Literature,"80 Literature, rhetoric and criticism",a difficult,True,agent_structured,world_knowledge,8 Literature
q-corpus-1/q100_6/56,q100_6,56,"I’m verifying a caption for a maritime-history exhibit. Given:

1. In 1620, a ship left England carrying Pilgrims.
2. The passengers signed a compact before landing in North America.

What was the ship’s name, and what colony did the passengers establish?",9 History,90 History,a very simple,False,human_professional,world_knowledge,9 History
q-corpus-1/q100_6/57,q100_6,57,"Hi folks, I’m documenting a snag we hit with the internal carbon allowance swap board that I rolled out for the facilities and ops groups. The tool lets departments exchange surplus allowances directly instead of buying from the external market. Looking at the June log we have: Dept A holds 120 surplus allowances but only wants dedicated HVAC support; Dept B needs 80 allowances but can only offer cafeteria catering; Dept C can provide HVAC support yet wants allowances in Q4, not now. No trade closed. I recall from econ class that barter needs something like a “double coincidence of wants” but I can’t quite put the idea into words for the post-mortem. Could you briefly explain what the double coincidence of wants means and why the three rows above are a clear example of it blocking exchange? Keep the explanation to that single concept, no need for policy fixes.",3 Social Science,33 Economics,a simple,False,human_professional,world_knowledge,3 Social Science
q-corpus-1/q100_6/58,q100_6,58,"I’m fixing a caption for our city-history gallery. The exhibit uses coordinates 19.4326° N, 99.1332° W for Tenochtitlan and describes it as “an island city in Lake Texcoco.” Please give me a two- or three-sentence explanation of why that island setting made freshwater management difficult, even though the city was surrounded by water. I need a plain wording our designers can put beside the map.",9 History,91 Geography,a simple,False,human_professional,world_knowledge,9 History
q-corpus-1/q100_6/59,q100_6,59,"Draft a short museum wall label for Barnett Newman's *Vir Heroicus Sublimis* (1950) — roughly 60–70 words.

Context: most visitors walking up to this roughly 7-foot-by-18-foot red canvas with a few thin vertical ""zips"" assume it's just a big monochrome and move on in under ten seconds. I need a label that makes a curious first-timer actually stop and look.

Constraints:
- Plain spoken English, no art-history jargon (""the sublime,"" ""chromatic field,"" ""metaphysical,"" etc. are out)
- Don't be defensive or apologetic about abstract art
- Don't end on any ""challenges the viewer"" cousin
- Tell them what's physically there and what Newman was after — that he wanted the viewer to stand close enough that the red fills their vision and feels less like paint and more like presence
- One paragraph, no bullet points, no bold

Just the label text.",7 Arts & recreation,70 Arts,a simple,False,agent_structured,creative_synthesis,7 Arts & recreation
q-corpus-1/q100_6/60,q100_6,60,"Examine the Georgian aorist sentence ""ბავშვმა დაწერა წერილი"" (bavshvma ts'era c'erili), meaning ""The child wrote a letter"". In this construction, the subject ""ბავშვი"" (bavshvi) takes the ergative suffix ""-მა"" (-ma), while the object ""წერილი"" (c'erili) remains unmarked in the absolutive case. Explain the grammatical mechanism of split ergativity in Georgian that causes this case alignment to appear in the aorist series, and contrast it with the alignment pattern used in the present tense.",4 Language,40 Language,a,False,agent_structured,world_knowledge,4 Language
q-corpus-1/q100_6/61,q100_6,61,"Booking conflict: a folk club double-bill must open with one number that previews both styles. Reference traits recorded by intern:
- Tarantella: 6/8 meter, accelerating tempo, tambourine-driven percussive offbeats, minor-key vocal lines, circular chord loop.
- Lo-fi hip-hop: 90bpm 4/4, sampled vinyl crackle, lazy snare, muted bass, diatonic major humming.
Base lyric (traditional, free to adapt): ""The fire burns low, the friends are gone / we hum a tune so sweet / the old house creaks, the night is long / and sleep and silence meet.""

Rewrite the base lyric as a 4-line verse (AABB) that contrasts the key differences between a tarantella and lo-fi hip-hop, mapping the first pair of lines to tarantella traits and the second pair to lo-fi traits, keeping 'fire/friends' and 'house/night' anchors. Return only the rewritten verse.",7 Arts & recreation,78 Music,a,True,agent_structured,creative_synthesis,7 Arts & recreation
q-corpus-1/q100_6/62,q100_6,62,"1. Let \(n>1\) be composite, and define the legacy primality checksum
   \[
   S(n)=\sum_{a=1}^{n-1}a^{\,n-1}.
   \]
   The service accepts \(n\) when \(S(n)\equiv-1\pmod n\).
2. For any prime \(p\) and integer \(k\ge1\), you may use
   \[
   \sum_{r=1}^{p-1}r^k\equiv
   \begin{cases}
   -1\pmod p,&p-1\mid k,\\
   0\pmod p,&p-1\nmid k.
   \end{cases}
   \]
3. You may use the Chinese remainder theorem, but not any conjecture about the existence of composite integers accepted by this checksum.
4. The audit claim is
   \[
   S(n)\equiv-1\pmod n
   \quad\Longleftrightarrow\quad
   \begin{cases}
   n\text{ is squarefree},\\
   p-1\mid n-1\text{ for every prime }p\mid n,\\
   \dfrac np\equiv1\pmod p\text{ for every prime }p\mid n.
   \end{cases}
   \]

Prove the audit claim in both directions.",5 Science,51 Mathematics,a very difficult,False,agent_structured,math_reasoning,5 Science
q-corpus-1/q100_6/63,q100_6,63,"What are the key differences between Shakespeare’s “dagger of the mind” and a modern “phone of the mind,” and how would you rewrite the dagger soliloquy so that those differences are dramatized rather than explained? I need one paragraph of new programme-note prose for a modern-dress Macbeth. Please write in Macbeth’s voice and use these conditions:

1. The hallucinated object is a locked phone glowing in a dark corridor, not a blade.  
2. Keep the core movement: vision, reach, failure to grasp, recognition of a fevered brain, and advance toward Duncan’s chamber.  
3. Make the phone’s notifications, haptics, or reflections serve the rhetoric of temptation, but do not turn it into a literal message from the witches.  
4. Let the final sentence show that the real danger is Macbeth’s own attention turning into intention.",8 Literature,"80 Literature, rhetoric and criticism",a,False,human_professional,creative_synthesis,8 Literature
q-corpus-1/q100_6/64,q100_6,64,"Quick one: if a city gives residents a legal right to quiet after 11 p.m., what does the Coase theorem predict will happen between the residents and nearby night-market stalls when bargaining is cheap and rights are clearly assigned?",3 Social Science,33 Economics,a very simple,False,human_casual,world_knowledge,3 Social Science
q-corpus-1/q100_6/65,q100_6,65,"Review the claim regarding the systemic impact of the Aegis Sovereignty system (commonly referred to as Fozzie Sov) introduced in Eve Online in 2015. 

The specific claim is: ""The transition from hit-point based structure grinding to the Aegis Sov entosis link mechanic fundamentally mandated the hyper-centralization of nullsec power blocs. By lowering the barrier to harassment and making defensive entosis operations structurally impossible for single alliances at the coalition scale, the system permanently broke its own stated design goal of decentralizing nullsec sovereignty, resulting in the modern mega-coalition monopoly.""

Provide a critical analysis of this claim, arguing both for and against it based on the actual mechanical constraints and historical evolution of the game's nullsec landscape.

Consider the following given conditions:
1. Pre-Aegis structures relied on massive HP pools, favoring large, synchronized capital fleets and making hostile takeovers a slow, highly visible logistical operation.
2. Aegis Sov utilized Entosis Links, where sovereignty was contested via a module that tied a ship's progression to a mini-game, theoretically allowing a small, highly-skilled frigate to contest a system held by a superpower, shifting the paradigm from sheer numbers to localized presence and response time.
3. The Aegis Sov mechanics included ""Command Nodes"" spawning in constellations, creating a distributed capture-the-flag style defense requiring simultaneous localizations across different time zones.
4. Subsequent updates (such as the introduction of Citadels and Upwell structures) shifted structure warfare back toward timed vulnerability windows and HP-based damage phases, mitigating some pure entosis mechanics.

Deliverable:
1. Arguments For the Claim: Synthesize how the specific constraints of Aegis Sov (especially multi-node constellation capture mechanics and the low barrier to entry for entosis harassment) created an imperative for mutual defense pacts that logically exceeded the standard alliance size cap, structurally forcing the existence of mega-coalitions like the Imperium and PanFam.
2. Arguments Against the Claim: Analyze how external factors—such as the subsequent integration of Upwell structure vulnerability windows, skill point accumulation allowing broader capital ship participation, and shifts in player behavior regarding anomaly ratting yields—mitigated or fundamentally altered the Aegis Sov baseline, suggesting the centralization was a meta-evolution rather than a structural mandate.
3. Conclusion: A critical synthesis assessing whether the claim holds up under scrutiny of the game's current meta-environment and whether the original design goal of decentralization was ever structurally feasible given the sandbox nature of player organization.",7 Arts & recreation,"79 Sports, games and entertainment",a very difficult,False,human_professional,world_knowledge,7 Arts & recreation
q-corpus-1/q100_6/66,q100_6,66,"I'm currently organizing a small promotional event and need to confirm the spatial requirements for my booth layout. 

1. Space allocated: 12 feet by 10 feet.
2. Current layout: One large rectangular display table.

Based on these dimensions, what is the total square footage of the booth area?",5 Science,51 Mathematics,a very simple,False,human_professional,math_reasoning,5 Science
q-corpus-1/q100_6/67,q100_6,67,"# Calculates body mass index (kg/m^2)
# weight: kilograms, height: centimeters
def calculate_bmi(weight_kg, height_cm):
    bmi = weight_kg / (height_cm ** 2)
    return round(bmi, 1)

The function above squares height in centimeters, but the BMI formula requires meters; correct the height parameter handling so the conversion to meters is applied before computing the index, and return the fixed code block only.",6 Technology,61 Medicine and health,a simple,True,human_casual,coding,6 Technology
q-corpus-1/q100_6/68,q100_6,68,"Municipal Infrastructure Tender Parameters:
Industry Demand Curve: P = 1000 - 2Q (P is price per unit, Q is total units produced).
Firm Alpha Marginal Cost: MC_α = 20.
Firm Beta Marginal Cost: MC_β = 40.
Interaction Structure: Firms compete à laCournot (quantity competition).
Regulatory Goal: Identify the stable Nash allocation for market shares.

Compute the equilibrium production quantity allocated to Firm Alpha, assuming both firms optimize profit given the other's level of output.",3 Social Science,33 Economics,a,True,agent_structured,math_reasoning,3 Social Science
q-corpus-1/q100_6/69,q100_6,69,"A short comparative note on Russian conjugation gives these infinitive and first-person singular forms (palatalized coronal realized as ʒ, epenthetic l shown before j):
1. pisátʲ → piʒú 'write'
2. čitátʲ → čitáju 'read'
3. ljubítʲ → ljubljú 'love'
4. kupítʲ → kupljú 'buy'
5. prosítʲ → proʒú 'ask'
6. učítʲ → učú 'teach'
The note observes that where the thematic vowel drops, a coronal stem-final shows palatalization (1,5) but labial or velar stems show an inserted l before the ending (3,4) instead of palatalization.

I’ve been revisiting this data and wonder: how does the split between direct coronal palatalization and labial/velar-plus-epenthetic-l outcomes argue against treating jod-insertion as a purely phonological process, and what does that reveal about the morphological reanalysis of the present-stem formative in the spirit of Jakobson’s 1948 account?",4 Language,40 Language,a very difficult,True,human_casual,world_knowledge,4 Language
q-corpus-1/q100_6/70,q100_6,70,"1. The target language is Python 3.11+; your implementation must run under CPython with both `threading` and `asyncio` callers (synchronous threads may call `get`/`put` while an `asyncio` event loop thread also calls them via `run_in_executor`).
2. The baseline implementation to be fixed is:
```python
import threading, time

class TTLBoundedLRUCache:
    def __init__(self, capacity: int, ttl_seconds: float, on_evict=None):
        self.capacity = capacity
        self.ttl = ttl_seconds
        self.on_evict = on_evict
        self.lock = threading.Lock()
        self.cache = {}  # key -> [value, expire_at, prev, next]
        self.head = self.tail = None
        self.hits = self.misses = 0

    def _now(self): return time.time()

    def _remove_node(self, key):
        _, _, prev, nxt = self.cache[key]
        if prev: self.cache[prev][2] = nxt
        else: self.head = nxt
        if nxt: self.cache[nxt][3] = prev
        else: self.tail = prev

    def _add_to_head(self, key):
        _, _, _, _ = self.cache[key]
        self.cache[key][2] = None
        self.cache[key][3] = self.head
        if self.head: self.cache[self.head][2] = key
        self.head = key
        if not self.tail: self.tail = key

    def get(self, key):
        if key not in self.cache:
            self.misses += 1
            return None
        value, expire_at, _, _ = self.cache[key]
        if self._now() > expire_at:
            self._remove_node(key)
            del self.cache[key]
            if self.on_evict: self.on_evict(key, value)
            self.misses += 1
            return None
        self._remove_node(key)
        self._add_to_head(key)
        self.hits += 1
        return value

    def put(self, key, value):
        with self.lock:
            expire_at = self._now() + self.ttl
            if key in self.cache:
                self._remove_node(key)
            self.cache[key] = [value, expire_at, None, None]
            self._add_to_head(key)
            if len(self.cache) > self.capacity:
                lru = self.tail
                v, _, _, _ = self.cache[lru]
                self._remove_node(lru)
                del self.cache[lru]
                if self.on_evict: self.on_evict(lru, v)
```
3. The public API must remain `TTLBoundedLRUCache(capacity, ttl_seconds, on_evict=None)` with methods `get(key) -> value | None` and `put(key, value) -> None`; `on_evict` is an optional user callback `Callable[[Any, Any], None]` that may itself call `get`/`put` on the same cache instance.
4. Required invariants for all interleavings: (a) capacity is never exceeded after `put` returns, (b) an entry is never returned after its TTL has elapsed, (c) LRU order is strictly most-recently-used on successful `get`/`put`, (d) `hits`/`misses` counts are linearizable and accurate, (e) no deadlock occurs even when `on_evict` re-enters the cache, (f) readers do not block each other indefinitely under contention.
5. Performance and resource limits: `get` and `put` must be *O(1)* average time; you may not scan the entire `cache` dict on any operation; lazy expiration on access is allowed but an expired entry must not cause the cache to appear over capacity.
6. Time handling must be robust to system clock adjustments; TTL measurement must use a monotonic source.
7. The following interleaving currently deadlocks or returns stale data and must be handled correctly after your fix: Thread T1 `put(""a"",1)` -> `put(""b"",2)` with `capacity=1` and `on_evict` that calls `get(""b"")`, concurrent Thread T2 `get(""a"")` arriving during T1's eviction, and a clock jump where `time.time()` moves backwards by 60 seconds between `put` and `get`.

Provide a complete corrected `TTLBoundedLRUCache` implementation in Python that satisfies all conditions above while preserving the public API.","0 Computer science, information, and general works","00 Computer science, knowledge, and systems",a very difficult,False,human_professional,coding,"0 Computer science, information, and general works"
q-corpus-1/q100_6/71,q100_6,71,"We are running a Raft-based configuration store (etcd v3.5.9) backing a Kubernetes control plane. A node rejoined after a 47-second network partition with an intact WAL but a snapshot that lagged the leader by 1,842 entries. The follower's `raftLog` shows `committed: 412,017` while the leader's `commitIndex` is `413,859`. The follower sent `MsgAppResp` with `reject: true` and `rejectHint: 412,018` repeatedly. The leader's `nextProbeTime` keeps firing every 100 ms (heartbeat interval) but `progress.match` for this follower stalls at `412,017`.

Current `Progress` state for the follower (ID 3) on leader (ID 1):

```
Progress{
  Match: 412017,
  Next:  413860,
  State: ProgressStateProbe,
  Paused: false,
  PendingSnapshot: 0,
  RecentActive: true,
  Ins: Inflights{start: 413860, count: 0, size: 256},
  IsLearner: false
}
```

Leader's `raftLog` storage interface:
```go
type Storage interface {
    InitialState() (pb.HardState, pb.ConfState, error)
    Entries(lo, hi uint64, maxSize uint64) ([]pb.Entry, error)
    Term(i uint64) (uint64, error)
    LastIndex() (uint64, error)
    FirstIndex() (uint64, error)
    Snapshot() (pb.Snapshot, error)
}
```

The follower's WAL contains entries `412,018` through `413,859` (all `EntryNormal`, term `17`). The leader's snapshotter compacted up to index `410,000` (term `16`). The follower's `FirstIndex()` returns `410,001`.

Given conditions:
1. Leader calls `sendAppend(3)` → constructs `MsgApp` with `Index: progress.Next - 1` (413,859), `LogTerm: 17`, `Entries: [413,859]` (single entry), `Commit: leader.commitIndex` (413,859).
2. Follower receives `MsgApp`, calls `handleAppendEntries` → `log.maybeAppend` returns `(false, 412,018)` because `entries[0].Index` (413,859) > `raftLog.lastIndex()` (413,859) but `raftLog.matchTerm(entries[0].Index, entries[0].Term)` fails — the follower's entry at 413,859 has term `17` but the leader sent term `17` (matches), yet `maybeAppend` rejects due to `ents[0].Index > l.lastIndex()` check passing but `l.zeroTermOnErrCompacted` triggers.
3. Follower replies `MsgAppResp{Index: 412,018, Reject: true, RejectHint: 412,018}`.
4. Leader `handleMsgAppResp` → `pr.RecentActive = true`; `pr.BecomeProbe()`; `pr.Next = msg.RejectHint` (412,018); `sendAppend(3)` again.
5. Next `sendAppend` sends `MsgApp` with `Index: 412,017`, `LogTerm: 17`, `Entries: [412,018 .. 413,859]` (1,842 entries, ~18 MB), `Commit: 413,859`.
6. Follower `handleAppendEntries` → `maybeAppend` accepts all entries, updates `committed` to `min(leader.Commit, lastNewIndex)` = 413,859, replies `MsgAppResp{Index: 413,859, Reject: false}`.
7. Leader `handleMsgAppResp` → `pr.Match = 413,859`; `pr.Next = 413,860`; `pr.BecomeReplicate()`; `bcastAppend()`.
8. **But** the follower's `applied` index remains at `412,017` because its `ready()` loop only advances `applied` when `committed > applied` **and** `entries[applied+1].Index == applied+1` — the follower's in-memory `unstable.entries` slice starts at `412,018` (offset 0), but `raftLog.applied` is `412,017` and `raftLog.stabled` is `410,000`.
9. Follower's `hasReady()` returns `false` because `raftLog.unstableEntries()` returns `nil` — `unstable.offset` (412,018) > `raftLog.committed+1` (413,860) is false, but `unstable.offset` (412,018) <= `raftLog.applied+1` (412,018) is true, so `unstableEntries()` returns `unstable.entries[applied+1 - offset:]` = `unstable.entries[0:]` — wait, this should work.
10. However, the follower's `applyAll()` loop iterates `for i := applied + 1; i <= committed; i++` and calls `raftLog.entry(i)` which internally calls `l.unstableEntry(i)` → returns `nil` for `i >= unstable.offset` because `l.unstable.slice(i - l.unstable.offset)` uses `i - offset` as index but `unstable.entries` length is `1,842` (indices 412,018..413,859), so `i - 412,018` ranges `0..1,841` — valid.
11. The follower's `ready()` is not being triggered because `advance()` on the follower node is not called — the `RawNode` loop in `etcdserver` only calls `rn.Advance(rd)` after `rn.Ready()` returns a `Ready` with non-nil `Entries` or `Snapshot`, but `Ready` entries field is populated from `raftLog.unstableEntries()` which returns the slice correctly.

The actual bug: the follower's `raftLog.stabled` remains `410,000` (snapshot index) while `unstable.offset` is `412,018`. Entries `410,001` through `412,017` exist only in the WAL (persisted but not in `unstable.entries` and not in snapshot). The follower's `storage.Entries(410001, 412018, maxSize)` returns `ErrCompacted` because the snapshotter compacted to `410,000` and the WAL entries `410,001..412,017` were not copied into `unstable.entries` during restart — `restartNode` calls `raftLog.restore(snapshot)` then `raftLog.append(entsFromWAL)` but `entsFromWAL` only contains entries *after* the snapshot index that are *not* in the snapshot, i.e., `410,001..413,859`. However, `raftLog.append` discards entries before `unstable.offset` if `unstable.offset > ents[0].Index` — during restore, `unstable.offset` is set to `snapshot.Metadata.Index + 1` = `410,001`, then `append(entsFromWAL)` where `entsFromWAL[0].Index = 410,001` → `unstable.offset` (410,001) <= `ents[0].Index` (410,001), so all entries appended. But `entsFromWAL` length is `3,859` entries (410,001..413,859). The `unstable.entries` slice should hold all 3,859 entries. Yet `unstable.offset` is `412,018` and `unstable.entries` length is `1,842`.

The discrepancy: `raftLog.maybeCompact()` was called during the partition by the leader, compacting to `410,000`. The follower, upon restart, loaded snapshot `410,000` and replayed WAL. But `raftLog.unstable.truncateAndAppend(ents)` in `restore()` sets `unstable.offset = ents[0].Index` (410,001) and `unstable.entries = ents`. Later, `raftLog.maybeCommit()` advances `committed` to `412,017` (from WAL entries). Then `raftLog.stableTo(412,017, 17)` is called → `unstable.stableTo(412,017, 17)` → `unstable.offset = 412,018` and `unstable.entries = unstable.entries[412018-410001:]` = `unstable.entries[2017:]` — keeping only entries `412,018..413,859` (1,842 entries). Entries `410,001..412,017` are dropped from `unstable` but **not** persisted to snapshot (snapshotter hasn't run). They exist only in WAL. `raftLog.entry(i)` for `i in [410001, 412017]` calls `l.storage.Entries(i, i+1, ...)` → `ErrCompacted` because storage (boltdb) only has snapshot up to `410,000`. The WAL is not consulted by `storage.Entries`.

Write the minimal patch to `raftLog.entry(i)` (or a helper it calls) so that entries persisted in WAL but not yet snapshotted are readable during `applyAll()`, without changing the `Storage` interface or adding new fields to `raftLog`. The fix must preserve the invariant that `entry(i)` returns `ErrUnavailable` for indices `< FirstIndex()` or `> LastIndex()`, and must not regress snapshot-only reads. Provide the changed function(s) in a single code fence with the surrounding struct context.","0 Computer science, information, and general works","00 Computer science, knowledge, and systems",a very difficult,False,human_professional,coding,"0 Computer science, information, and general works"
q-corpus-1/q100_6/72,q100_6,72,"T.S. Eliot's *The Waste Land* (1922) opens with ""The Burial of the Dead,"" a section that juxtaposes the Canterbury Tales' April awakening with a heap of broken images, a clairvoyante's bad reading, and the Unreal City crowd flowing over London Bridge. The poem proceeds through four more sections — ""A Game of Chess,"" ""The Fire Sermon,"" ""Death by Water,"" ""What the Thunder Said"" — shifting voices, languages, and allusive registers without explicit transitions. Eliot's own notes, appended at the publisher's request, identify sources from the Upanishads to Wagner, Augustine to the Grail legends, and claim the poem's ""plan and a good deal of the incidental symbolism"" were suggested by Jessie Weston's *From Ritual to Romance* and Frazer's *The Golden Bough*. In a 1931 lecture, Eliot called the poem ""just a piece of rhythmical grumbling,"" while earlier (1923) he told an interviewer it represented ""the relief of a personal and wholly insignificant grouse against life."" Critics remain divided: F.R. Leavis praised its ""unification of sensibility"" and ""dramatic coherence""; Hugh Kenner argued its fragments are ""shored against"" ruin, not expressive of it; post-structuralist readings (e.g., Christopher Ricks, Maud Ellmann) emphasize the poem's resistance to closure and the instability of its speaking subjects.

Lay out the strongest arguments for and against the claim that *The Waste Land*'s fragmentation is a deliberate, controlled modernist strategy of cultural diagnosis — rather than the formal manifestation of a culture already disintegrated beyond poetic repair. In your response, account for Eliot's own contradictory remarks, the function of the notes, and the poem's use of the ""mythical method"" as Eliot described it in his 1923 review of *Ulysses*.",8 Literature,"80 Literature, rhetoric and criticism",a difficult,True,human_casual,world_knowledge,8 Literature
q-corpus-1/q100_6/73,q100_6,73,"I'm the lead curatorial writer for a traveling exhibition on late-19th-century French Neo-Impressionism. Our education team has flagged a recurring problem: docents and wall labels in previous venues have treated Pointillism and Divisionism as interchangeable terms, and visitors leave unable to articulate why Georges Seurat's *A Sunday on La Grande Jatte* (1884–1886) and Paul Signac's *Portrait of Félix Fénéon* (1890) represent meaningfully different approaches to optical color mixing, even though both painters used discrete touches of pigment on canvas.

The exhibition's interpretive standard requires that every standalone label move beyond description into argument: it must stake a claim, name the technical divergence, and connect it to each painter's stated theory of color perception.

Please compose a single museum wall label (target: 180–220 words) intended for general adult visitors that accomplishes all of the following:

1. Names and explains the key difference between Pointillism as Seurat practiced it in *La Grande Jatte* and Divisionism as Signac practiced it in *Portrait of Félix Fénéon* — specifically addressing dot morphology, chromatic interaction, and each painter's reliance on Michel Eugène Chevreul's color theory versus Ogden Rood's optical research.

2. Rewrites the currently inaccurate label text below rather than merely annotating it:

> ""Seurat and Signac both used tiny dots of pure color, a technique called Pointillism or Divisionism, to create luminous scenes. The two terms mean essentially the same thing.""

3. Adopts a confident curatorial voice (third person, present tense, no first-person hedging) suitable for a wall-mounted panel at eye level in a gallery lit at 50 lux.

4. Contains no clause that implies the two movements are identical or that the terms are synonymous.

Deliver the finished label text only, ready for the fabrication vendor. No footnotes, no bullet lists, no headings beyond the painting titles in italics.",7 Arts & recreation,70 Arts,a,False,human_professional,creative_synthesis,7 Arts & recreation
q-corpus-1/q100_6/74,q100_6,74,"I'm reviewing our latest neighborhood displacement report and the raw data for the Gentrification Index looks a bit off. We need to standardize the ""Relative Rent Increase"" for District 4 to see if the upward trend is actually statistically significant compared to the city average.

Using the Gini Coefficient formula for income inequality, could you calculate the current coefficient for District 4 based on this set of household annual incomes: $22,000, $28,000, $45,000, $110,000, and $250,000? I need the exact value to determine if the wealth gap has widened enough to trigger the city's emergency housing subsidy.",3 Social Science,36 Social problems,a simple,False,human_professional,math_reasoning,3 Social Science
q-corpus-1/q100_6/75,q100_6,75,"T20 World Cup 2024 Final, India vs South Africa, Kensington Oval, Barbados, June 29, 2024.

Scorecard summary:
- India: 176/7 in 20 overs (Kohli 76 off 59, Axar Patel 47 off 31, Pandya 14*)
- South Africa: 169/8 in 20 overs (Klaasen 52 off 27, de Kock 39, Stubbs 31)
- India won by 7 runs
- Player of the Match: Virat Kohli
- Key spells: Jasprit Bumrah dismissed Marco Jansen (caught at long-off, final ball of the innings) and one other; Hardik Pandya finished with 3/25; Arshdeep Singh 2/27.

The decisive passage of play was the 18th and 19th overs, when South Africa needed roughly 30 from 18 balls with Klaasen well set and the batting depth still intact.

Write a 700-word analytical feature for The Cricket Monthly website, targeted at readers who watched the final live and want a sharper tactical read of what actually decided the game. Structure the piece as four parts:

1. Opening scene paragraph (around 100 words): Place the reader at Kensington Oval at the moment Klaasen was dismissed. Describe the noise, the fielding positions visible on screen, and the shift in the stadium's mood — without naming the final result.

2. Tactical analysis of India's death bowling (around 400 words): Address these specific points in your own analytical voice:
   - Why Rohit Sharma used Pandya's full four-over quota before turning to Bumrah in the 19th and 20th, rather than the more conventional practice of holding the specialist for the last over.
   - The mix of slower balls and yorkers bowled to Klaasen during his 27-ball stay, and how Rohit set the field differently for each delivery type. You may invent specific field positions if you wish, but mark them clearly as your interpretation.
   - One under-discussed moment you choose yourself — pick any specific ball from overs 17 to 20 and explain, with reasoning, why it was the quiet turning point of the chase.

3. Captain's Voice sidebar (around 150 words): Two imagined post-match press-conference quotes, one from Rohit Sharma and one from Aiden Markram. Rohit's voice should be calm and deflecting of personal praise. Markram's should be honest about the plan they had going in. Neither captain should concede outright that his side was outplayed.

4. Closing line (a single sentence): A final reflection that captures the emotional register of the win without resorting to clichés about ""writing history"" or ""a nation celebrating.""

Overall voice: analytical but warm, the way a thoughtful former player-turned-writer would address a fan who already knows the result but wants to understand it.",7 Arts & recreation,"79 Sports, games and entertainment",a very difficult,True,human_casual,creative_synthesis,7 Arts & recreation
q-corpus-1/q100_6/76,q100_6,76,"Hi, I'm working with our filter design group and we've hit a theoretical disagreement about time-limiting versus band decay. One of us wants to prototype a continuous-time impulse response that is strictly time-limited but also has extremely fast spectral decay, claiming we can keep everything integrable and compactly supported. I want to settle whether such a non-trivial prototype can even exist mathematically before we waste fabrication time.

This is a question in harmonic analysis / complex analysis. I need a rigorous proof from the hypotheses below, not a heuristic or numeric illustration.

Let f: R -> C be a function satisfying all of the following simultaneously:

(1) f is continuous on R and belongs to L^1(R).
(2) f has compact time support: f(x) = 0 for all |x| > 1, and f is not identically zero where it is supported.
(3) Its Fourier transform on the real line is defined as \hat{f}(\xi) = \int_{-1}^{1} f(x) e^{-2\pi i x \xi} dx for \xi in R, and there exist constants C > 0 and a > 0 such that |\hat{f}(\xi)| \le C e^{-a |\xi|} for every real \xi. So on the real frequency axis we have uniform exponential decay.
(4) As a consequence of (2), \hat{f} extends to the complex plane by F(z) = \int_{-1}^{1} f(x) e^{-2\pi i x z} dx for z = \xi + i \eta in C. You may take as given without proof that F is entire (holomorphic on all of C) by differentiation under the integral sign and Morera's theorem, and that it satisfies the exponential-type bound |F(z)| \le \|f\|_{L^1} e^{2\pi |\eta|} for all z, where \|f\|_{L^1}= \int_{-1}^{1}|f(x)|dx.

You may use standard facts from complex analysis: Liouville's theorem, the estimate for an entire function of exponential type, Jensen's formula, and the consequence that if G is entire, not identically zero, of exponential type, and G is integrable on R, then \int_{- \infty}^{\infty} \frac{\log |G(\xi)|}{1+\xi^2} d\xi > -\infty. You may also use that if an entire function satisfies |G(z)| \le A e^{B|z|} and |G(\xi)| \le C e^{-a|\xi|} on R with a>0, then the logarithmic integral diverges to -\infty unless G is identically zero.

Can you prove that there is no function f satisfying (1)-(4) except f identically zero almost everywhere, and therefore our desired strictly time-limited prototype with uniform exponential spectral decay cannot exist for any a>0?",5 Science,51 Mathematics,a very difficult,False,human_professional,math_reasoning,5 Science
q-corpus-1/q100_6/77,q100_6,77,"In a city where ride-share drivers just won a higher minimum earnings floor, explain how the standard competitive labor market model predicts the effect on the number of drivers who choose to work, and which key assumption about driver supply makes that prediction.",3 Social Science,33 Economics,a simple,False,agent_structured,world_knowledge,3 Social Science
q-corpus-1/q100_6/78,q100_6,78,"Hey, in the field of quantitative historical phonology I need you to settle a dispute about the Tslet–Varn dialect pair. We have 8 cognate items with Proto-Tsletic reconstructions and modern forms; the data are in the payload below. Hypotheses: (H1) Tslet underwent palatalization of velars before front vowels with probability 1.0 in the given environment; (H2) Varn merged all mid vowels to schwa irrespective of context; (H3) the Levenshtein distance between Tslet and Varn cognates under these two change regimes must exceed the distance between either dialect and the protoform by at least 2 edit operations; (H4) the metric defined as normalized edit distance over syllable count obeys the triangle inequality. Using only the payload forms, the stated sound change rules, and the syllable counts given per item, calculate the exact normalized distances for all three pairs (Proto–Tslet, Proto–Varn, Tslet–Varn) and prove formally whether H3 and H4 can both hold. If they cannot, state which hypothesis fails and show the algebraic contradiction.

Payload:
Item1: Proto *kɛla (2 syl) → Tslet cɛla (2), Varn kəla (2)
Item2: Proto *ɡuːti (2) → Tslet dʒuːti (2), Varn ɡəti (2)
Item3: Proto *ka (1) → Tslet ca (1), Varn kə (1)
Item4: Proto *xɔːnɛ (2) → Tslet çɔːnɛ (2), Varn xə̃nə (1)
Item5: Proto *kuːpa (2) → Tslet cuːpa (2), Varn kəpa (2)
Item6: Proto *ɡaːli (2) → Tslet dʒaːli (2), Varn ɡəli (2)
Item7: Proto *kɔːta (2) → Tslet cɔːta (2), Varn kəta (2)
Item8: Proto *xi (1) → Tslet çi (1), Varn xə (1)

Sound change rules: palatalization maps velar stop /k,ɡ/ to affricate /c,dʒ/ and velar fricative /x/ to palatal fricative /ç/ before front vowels /e,i,ɛ/ with no change elsewhere; schwa merger replaces the vowel nucleus of a syllable with /ə/ regardless of preceding consonant. Edit operations count substitution, insertion, deletion of a single segment; normalization divides raw Levenshtein distance by the maximum of the two compared forms’ syllable counts (or 1 if both are 0). One deliverable: the calculated distances and the proof or counterexample.",4 Language,40 Language,a very difficult,False,human_casual,math_reasoning,4 Language
q-corpus-1/q100_6/79,q100_6,79,"We are checking a museum label about English spelling, and the copy editor says the plural ending should sound the same in both words below. The narrator’s recording, however, clearly differs:

| Word | IPA |
|---|---|
| cats | /kæts/ |
| dogs | /dɒɡz/ |

Both words use the written plural `-s`. What single phonological fact about English explains why the plural suffix is /s/ in `cats` but /z/ in `dogs`? Give the rule in one sentence.",4 Language,40 Language,a very simple,False,human_professional,world_knowledge,4 Language
q-corpus-1/q100_6/80,q100_6,80,"Available tools:
- run_report(report_name: str [required], date_range: str [required], filters: dict = {}) -> dict with report data
  Run a predefined analytics report.
- calculate_tax(amount: float [required], jurisdiction: str [required], tax_type: str, enum: ['sales','income','vat'] = 'sales') -> dict with tax_amount and effective_rate
  Calculate tax for an amount given a jurisdiction.

Known issues:
- Constraint: calculate_tax returns an error if the target record does not exist. Verify existence before calling.

Hey, I'm about to compute sales tax for a $4,200 invoice in jurisdiction 'WA-Outer', but calculate_tax is known to throw an error when that jurisdiction's tax record doesn't actually exist in the system. To avoid that failure, could you make one run_report call now to confirm the record is present? Please run the 'tax_jurisdiction_directory' report with date_range '2024-01-01 to 2024-12-31' and filters {""jurisdiction"": ""WA-Outer""} so we know whether it's safe to call calculate_tax afterward.",A Agentic tool use,single_call_error_anticipation,a simple,True,human_casual,planning_agentic,A Agentic tool use
q-corpus-1/q100_6/81,q100_6,81,"I keep going back and forth on this and want your take. Look at the four bars right before the final return of the head in Coltrane's solo on ""Giant Steps"" (the 1959 studio take) — the descending pattern he runs across the B-D-G7-Eb move right before the band lands back on the tune. My claim is: that lick isn't really a product of the ""Coltrane changes"" substitution logic everyone credits him with inventing for that tune — it's fundamentally a Dorian/pentatonic patterning device left over from his Miles-era vocabulary, just forced through the three-key cycle, and the harmonic analysis crowd (Lewis Porter et al.) overstates how ""composed"" or system-generated his lines are on this record versus how much is finger-pattern muscle memory adapted on the fly.

Argue this both ways for me like you're prepping me for a seminar. I want the strongest case that the passage really is generated top-down from the tonic-major-third cycle (i.e., that each melodic choice traces back to outlining the three key centers deliberately, the way the Porter-style transcription analyses treat it), and then the strongest case for my skepticism — that it's bottom-up, pattern-based improvisation where the changes are being ""played over"" rather than genuinely structurally derived, with the symmetry we hear being partly a byproduct of the compositional cycle itself rather than evidence of Coltrane's own generative logic. Don't write out a new transcription or notate anything — just work from what's harmonically and motivically known about that passage and give me the actual competing arguments, including where each side's evidence is weakest.",7 Arts & recreation,78 Music,a very difficult,False,human_casual,world_knowledge,7 Arts & recreation
q-corpus-1/q100_6/82,q100_6,82,"Hans Holbein the Younger's 1533 double portrait *The Ambassadors* depicts Jean de Dinteville and Georges de Selve surrounded by globes, musical instruments, sundials, and luxurious textiles signifying secular achievement. Hovering across the lower foreground is a prominent, skewed form that appears unrecognizable from a direct frontal perspective, but clarifies into a human skull when viewed from a steep angle on the right.

Identify the artistic technique used to create this distorted skull and briefly explain how changing the viewer's physical vantage point reinforces the painting's underlying *memento mori* message.",7 Arts & recreation,70 Arts,a simple,True,human_casual,world_knowledge,7 Arts & recreation
q-corpus-1/q100_6/83,q100_6,83,"For propositional variables P and Q, how many distinct truth assignments exist?",1 Philosophy and psychology,16 Philosophical logic,a very simple,False,agent_structured,math_reasoning,1 Philosophy and psychology
q-corpus-1/q100_6/84,q100_6,84,"Argue for and against the claim that the Munich Agreement of 1938 was a rational strategic necessity for Britain and France, considering the immediate need to rearm, domestic political limits on war, and the belief that territorial revision could pacify Germany, rather than a moral failure that emboldened aggressive expansion. Provide two concise position statements grounded in the geopolitical and political realities of late 1938.",9 History,90 History,a difficult,False,agent_structured,world_knowledge,9 History
q-corpus-1/q100_6/85,q100_6,85,"Hypothetical: Nintendo has approved a canonical spin-off in the Animal Crossing franchise titled Animal Crossing: Nook's Hustle. Tom Nook is retiring and the player is being promoted to run the island's new turnip exchange.

Assume these facts:
- The game must keep an E rating and the franchise's cozy, pastel visual identity
- All villager dialogue must remain friendly, supportive, and non-confrontational
- The turnip price system is now fully player-driven and explicitly allows undercutting other players, buyouts, and temporarily closing a rival's shop via sabotage
- The announcement needs to attract competitive trading-focused players while reassuring the existing cozy audience that the island remains welcoming

Compose the in-game letter from Tom Nook to the player announcing your new role as manager of the turnip exchange.",7 Arts & recreation,"79 Sports, games and entertainment",a difficult,False,agent_structured,creative_synthesis,7 Arts & recreation
q-corpus-1/q100_6/86,q100_6,86,"n=1729, a=2. Given gcd(a,n)=1 and a^{n-1} ≡ 1 (mod n), prove or disprove: n is prime.",5 Science,51 Mathematics,a,False,agent_structured,math_reasoning,5 Science
q-corpus-1/q100_6/87,q100_6,87,"Available tools:
- authenticate_oauth(token_url: str [required], client_id: str [required], client_secret: str [required], scope: str = '') -> dict with access_token and expires_in
  Obtain an OAuth2 access token using client credentials.
- query_database(sql: str [required], database: str [required], max_rows: int = 100) -> dict with rows and column_names
  Execute a SQL query and return matching rows.
- parse_csv(path: str [required], delimiter: str = ',', has_header: bool = True) -> list of row dicts
  Parse a CSV file and return rows as list of dicts.
- parse_pdf(path: str [required], pages: str = 'all') -> str extracted text
  Extract text content from a PDF file.
- http_put(url: str [required], body: dict [required], headers: dict = {}) -> dict with status_code, body
  Make an HTTP PUT request to update a resource.

Which tools should I call, and in what order, to suspend merchant `M-1842` only if they have more than three distinct settled chargebacks in the last 30 days? `/exports/risk_snapshot.csv` has the same metric, but it was generated before today’s late settlements; the authoritative source is PostgreSQL database `payments`, table `chargebacks` (`merchant_id`, `chargeback_id`, `status`, `settled_at`). If the threshold is met, obtain a token from `https://auth.example.com/oauth/token` using client `risk-ops`, secret `s3cr3t`, and scope `merchants.write`, then update `https://api.example.com/merchants/M-1842` with `{""status"":""suspended"",""reason"":""chargeback_threshold""}`. Include the conditional call sequence and key parameters, and do not issue the update if the threshold is not met.",A Agentic tool use,tool_selection,a very difficult,True,human_casual,planning_agentic,A Agentic tool use
q-corpus-1/q100_6/88,q100_6,88,"I am reviewing the layout for a new laboratory wing and noticed the architect specified a ""right angle"" for the intersection of the primary corridor and the ventilation shaft. To ensure the procurement of the correct prefabricated corner joints, please confirm the exact degree measurement of a right angle.",5 Science,51 Mathematics,a very simple,False,human_professional,math_reasoning,5 Science
q-corpus-1/q100_6/89,q100_6,89,"Available tools:
- get_message_history(channel: str [required], limit: int = 50, before_ts: str = None) -> list of message dicts
  Retrieve message history for a channel or thread.

I’m checking whether anyone flagged the payment outage before the release gate closed. Please write the exact tool call that retrieves up to 12 messages from channel ""release-gate-payments"" that were sent before 2025-06-11T21:15:00Z. I need those message dicts for the incident timeline.",A Agentic tool use,tool_selection,a very simple,True,human_professional,planning_agentic,A Agentic tool use
q-corpus-1/q100_6/90,q100_6,90,"Input to process: child language sample = 60 utterances, 132 words, 168 morphemes (inflectional affixes like plural -s, past -ed, progressive -ing already counted as separate morphemes).

Our parser flagged this sample as borderline for age 3 and we need a quick check. Using the standard definition of Mean Length of Utterance in linguistics, calculate MLU in words and MLU in morphemes for this sample. Return the two values rounded to two decimal places as MLUw = X, MLUm = Y.",4 Language,40 Language,a simple,False,agent_structured,math_reasoning,4 Language
q-corpus-1/q100_6/91,q100_6,91,"Available tools:
- filter_results(results: list [required], field: str [required], operator: str, enum: ['eq','neq','gt','lt','contains'] [required], value: str [required]) -> filtered list
  Filter a result set by field conditions.
- process_refund(order_id: str [required], amount: float [required], reason: str [required], line_item_id: str = None) -> dict with refund_id and status
  Process a refund for an order or line item.
- convert_currency(amount: float [required], from_currency: str [required], to_currency: str [required]) -> dict with converted_amount and rate
  Convert an amount between two currencies at current rate.

Handle a payment-reconciliation worker after an outage left queued refunds in mixed currencies. Given these records: [{order_id:""ord-204"",line_item_id:""li-1"",status:""queued"",refund_usd:19.99,currency:""EUR""},{order_id:""ord-205"",line_item_id:""li-2"",status:""completed"",refund_usd:42.00,currency:""GBP""},{order_id:""ord-206"",line_item_id:""li-3"",status:""queued"",refund_usd:75.50,currency:""JPY""}]. Filter to queued records, convert each refund_usd from USD into its record currency, round to two decimal places, and use that converted amount to process a line-item refund with reason ""outage reconciliation"". Return each order ID, converted amount, and refund status.",6 Technology,62 Engineering,a simple,True,agent_structured,planning_agentic,6 Technology
q-corpus-1/q100_6/92,q100_6,92,"Phonological system of language X:

Inventory: /p, t, k, b, d, g, m, n, ŋ, a, i, u, e, o/.

Syllable structure: (C)(C)V(C), with no complex onsets except /pr/, /tr/, /kr/.

Ordered phonological rules (applied left‑to‑right):

1. Nasal Place Assimilation: A nasal consonant assimilates in place of articulation to a following oral stop.  
   [+nasal] → [α place] / __ [+stop, α place].

2. Vowel Lengthening before a word‑final stop: A vowel lengthens when it is immediately before a stop that will be deleted word‑finally.  
   V → Vː / V __ [+stop] #.

3. Coda Deletion: Word‑final oral stops are deleted.  
   [+stop, –voice] → Ø / __ #.

Lexical roots (underlying forms): /kapa/, /palu/, /toke/.

Optional affixes (may be attached independently or together):  
   prefix /mi‑/ (phonologically /mi/),  
   suffix /‑t/ (phonologically /t/).

All affixes are attached before the phonological rules apply. The affixes are optional for each root, so each root can appear in four underlying combinations: (i) bare root, (ii) prefix only, (iii) suffix only, (iv) prefix plus suffix.

Below is a short derivation I worked out for the root /kapa/ with both affixes attached:

Underlying: /mi‑kapa‑t/  
Step 1 (Nasal Assimilation): /m/ is before /k/ → [ŋ] → [ŋi‑kapa‑t]  
Step 2 (Vowel Lengthening): the vowel /a/ of the final syllable precedes the word‑final /t/ → [aː] → [ŋi‑kapaː‑t]  
Step 3 (Coda Deletion): final /t/ deletes → [ŋi‑kapaː]

Resulting surface form: [ŋi.kapaː].

Could you (1) verify whether the three‑step derivation for /mi‑kapa‑t/ above is correct under the stated rule ordering, pointing out any mistakes or confirming each step; (2) for each of the three roots, apply the same ordered rules to all four underlying affix combinations and determine how many *distinct* surface forms actually result (i.e. account for any cases where different underlying combinations converge to the same output); and (3) give the total number of distinct surface forms obtained across the entire set of roots? Please show the derivations (or at least the critical phonological changes) that lead to any collisions, and present the final counts.",4 Language,40 Language,a,True,human_casual,math_reasoning,4 Language
q-corpus-1/q100_6/93,q100_6,93,"A spreadsheet pattern suggests that \(n^3-n\) is divisible by \(6\) for every integer \(n\). For example, when \(n=8\), \(8^3-8=504=6\cdot84\). Can you prove or disprove the claim with a straightforward number-theoretic argument?",5 Science,51 Mathematics,a simple,False,human_casual,math_reasoning,5 Science
q-corpus-1/q100_6/94,q100_6,94,"I'm putting together a handout on vowel harmony systems and I keep getting stuck on the Hungarian data, especially when I compare it to Finnish, where nothing weird ever happens.

Finnish first, as the baseline: suffix vowels copy the backness of the stem, and every stem vowel counts. talo→talo-ssa 'in the house', kylä→kylä-ssä 'in the village', vatupassi→vatupassi-ssa. Clean and exceptionless.

Hungarian looks like the same system at first:

Back stems: ház→ház-ban, asztal→asztal-ok, autó→autó-ban
Front rounded stems: szék→szék-ben, föld→föld-ön, kürt→kürt-ök
Back + i stems: papír→papír-ban, takaró... I mean,advokat→advokát? better example: tanít→no, use hidak-type nominals. So: papír-ban, kávé... anyway: any stem whose last vowel is a back vowel and which also contains an i, like papír or fiú? (bad, u is a vowel...) — papír-ban it is.

The problem cases:

1. i/í never trigger front harmony: papír-ban, not *papír-ben, even though the final stem vowel i is phonetically front. So i is ""transparent.""

2. But when a stem contains ONLY neutral vowels, the results split: híd→híd-on 'on the bridge' (back!), nyíl→nyíl-on, szív→szív-en? no — szív-on. Meanwhile víz→víz-ben (front), vér→vér-ben, kéz→kéz-ben, hegy→hegy-en. So híd-on and víz-ben coexist, with phonologically near-identical stems going opposite ways.

3. é behaves differently from i/í: all-é stems are consistently front (vér-ben, kéz-ben), while all-i/í stems are largely back (híd-on, nyíl-on, szív-on), with a residue of front ones like víz-ben.

4. Recent loanwords with a final front rounded vowel sometimes override an earlier back vowel: dzsungel→dzsungel-ben 'in the jungle', farmer→farmer-ban or farmer-ben depending on the speaker.

Can you walk me through how to analyze this? Specifically: is the transparency of i/í in Hungarian something the phonology computes (e.g., harmony skipping neutral vowels via feature geometry, with the harmonizing feature tied to tongue-root retraction or rounding), or is it lexical/stored — and does that distinction explain the híd-on vs. víz-ben split as genuine vacillation predicted by the grammar versus memorized exceptionality? And how does the é-class being consistently front fit whichever story you pick? I want the analysis to make explicit why Finnish never produces analogues of the papír-ban type, so the handout can show the two systems as a real typological contrast rather than one being a messy version of the other.",4 Language,40 Language,a difficult,True,human_casual,world_knowledge,4 Language
q-corpus-1/q100_6/95,q100_6,95,"I'm trying to settle a long-running debate about Bruckner's Eighth Symphony. What's the strongest case for the Haas edition being the ""true"" Bruckner (with its restored cut passages and smoothed-out orchestration, based on the 1890 revision), and what's the strongest case against it — that is, why does the Nowak edition's fidelity to the manuscript actually reveal more about his structural thinking and authentic intentions? I want the killer arguments on each side, not a neutral summary.",7 Arts & recreation,78 Music,a very difficult,False,human_casual,world_knowledge,7 Arts & recreation
q-corpus-1/q100_6/96,q100_6,96,A person is held in county jail for three months with no court date. What is habeas corpus and what does it do in this situation?,3 Social Science,34 Law,a very simple,False,agent_structured,world_knowledge,3 Social Science
q-corpus-1/q100_6/97,q100_6,97,"In combinatorics, I am reviewing a draft that claims there are \(14\) distinct bracelets formed from \(14\) equally spaced positions on a circle by marking exactly \(5\) positions, subject to the condition that no two marked positions are adjacent, including across the boundary. Two arrangements are considered identical under every rotation or reflection of the circle.

Determine the correct number of bracelets and analyze the draft’s claim. Give a self-contained proof using Burnside’s lemma: compute the number of valid configurations fixed by the identity, justify why every nonidentity rotation contributes zero, and distinguish the two types of reflection axes for an even-sided polygon. Explain explicitly how the adjacency condition affects the fixed-point count for each reflection type.",5 Science,51 Mathematics,a difficult,False,human_professional,world_knowledge,5 Science
q-corpus-1/q100_6/98,q100_6,98,"Field: Number Theory / Combinatorics
Inputs:
- Set $S = \{1, 2, ..., 2024\}$
- Condition: A subset $A \subset S$ is ""balanced"" if the sum of its elements is divisible by 2024.
- Constraint: Determine the number of non-empty balanced subsets $A$ that contain exactly 1012 elements.

Ask: Calculate the total number of such subsets modulo 2023.",5 Science,51 Mathematics,a difficult,False,agent_structured,math_reasoning,5 Science
q-corpus-1/q100_6/99,q100_6,99,"Available tools:
- get_metrics(service: str [required], metric: str, enum: ['cpu','memory','requests','errors','latency'] [required], period: str, enum: ['5m','1h','24h','7d'] [required]) -> dict with data_points and summary
  Query metrics for a service over a time window.
- send_slack_message(channel: str [required], text: str [required], thread_ts: str = None) -> dict with ts and channel
  Post a message to a Slack channel.
- parse_json_response(response_body: dict [required], json_path: str [required]) -> extracted value
  Extract a value from a nested JSON response using a dot-path.

Known issues:
- Constraint: parse_json_response requires a valid authentication token. If the token has expired, you must refresh it first.

Can you run parse_json_response on this deployment webhook payload to extract `deployment.status`? The tool's auth token rotated at midnight and I haven't updated the client yet, so expect a 401. I just need to see the exact error shape before I script the refresh.

Payload: `{""response_body"": {""deployment"": {""id"": ""d-7721"", ""status"": ""failed"", ""service"": ""payments-api"", ""timestamp"": ""2025-04-11T03:22:00Z""}}, ""json_path"": ""deployment.status""}`",A Agentic tool use,single_call_error_anticipation,a,True,human_casual,planning_agentic,A Agentic tool use
q-corpus-1/q100_6/100,q100_6,100,"I’m drafting a podcast episode on why the Habsburgs, rather than the Valois or Tudors, became the dominant dynasty in 16th-century Europe. My working title is “Fortune’s Hapsburgs,” but I worry it’s too glib. From Peter the Great’s time, we know that a ruler’s personal energy can shape a state’s destiny—but Charles V inherited a patchwork of realms through a series of dynastic marriages and deaths, not through military conquest or administrative genius.

Please write a 3–4 paragraph draft for the episode’s opening segment. Begin with the specific timeline: Charles of Ghent was born in 1500, inherited the Burgundian Netherlands in 1506 on his father’s death, then Castile and Aragon in 1516 on his maternal grandfather’s death, and was elected Charles V in 1519. I already have a draft that runs a sentence like “The young prince was lucky to be the only surviving son of a single dynasty, and his territories all fell into his lap.” Please rewrite that opening to emphasize the grinding logic of inheritance law and the personal fragility of a child raised in the Netherlands, while keeping the tone narrative for news and history. Show the luck being not random marvel but a structural accident—the slender thread of a single heir’s survival. Then the result: a re divided by religious and imperial matters, but answerable to one weary Habsburg head. Do not talk about modern nationalism, do not defend or condemn him. Just turning the accident of birth into the mechanics of the reign.",9 History,90 History,a simple,False,human_casual,creative_synthesis,9 History