Prefill-Free Cross-Family KV Cache Transfer for Heterogeneous Multi-Agent LLMs
Paper • 2609.32259 • Published • 102
Owen Reed
owenreed
·
AI & ML interests
mostly healthcare AI, unit test generation, making models smaller, batching strategies, efficient inference, FlashAttention
Recent Activity
upvoted a paper 8 days ago
Prefill-Free Cross-Family KV Cache Transfer for Heterogeneous Multi-Agent LLMs upvoted a paper 8 days ago
Video Generation Models: A Survey of Post-Training and Alignment upvoted a paper 20 days ago
Harness-Zero: Harness Distillation via Agent-as-HarnessOrganizations
None yet
upvoted 2
papers 8 days ago
upvoted 2
papers 20 days ago
upvoted 2
papers 22 days ago
upvoted 2
papers 23 days ago
upvoted a paper 25 days ago
upvoted 3
papers 26 days ago
Grouped Value Attention: Efficient KV Caching via On-Demand Key Reconstruction
Paper • 2609.13285 • Published • 82
How Lossless Is Lossless Speculative Decoding? The Role of Numerical Precision in Orthrus
Paper • 2609.15504 • Published • 37
Atria Dawn: The Dawn of Agentic Superintelligence
Paper • 2609.15818 • Published • 410
upvoted a paper 28 days ago
upvoted a paper 30 days ago
upvoted 6
papers about 1 month ago
BeaconKV: Key-Value Cache Compression Guided by Beacon Queries for Efficient Large Reasoning Model Inference
Paper • 2609.04971 • Published • 43
Language Models Can Control Their Own Attention
Paper • 2609.02737 • Published • 79
Unlocking Lossless Speedups in LLMs via Discrete Diffusion
Paper • 2609.04010 • Published • 114
Online Draft Co-Training for Speculative Decoding in Large-Scale, Long-Context RL Post-Training
Paper • 2609.07108 • Published • 36
NeoHorse-1: Towards Recursive Self-Improvement via Agentic Post-Training with Routing Harness
Paper • 2609.08183 • Published • 321
Random Attention: Rethinking KV Cache Eviction for Efficient Reasoning
Paper • 2609.03430 • Published • 188