Post
47
Real Long-Term Memory for AI: A 50-Million-Token Window That Is Faster and Cheaper Than Recompute https://www.reddit.com/r/huggingface/s/T2tXxKlqSb
AI & ML interests
None defined yet.
Recent Activity
Papers
Real Long-Term Memory for AI: A 50-Million-Token Window That Is Faster and Cheaper Than Recompute
Working Around the Compute Ceiling: Byte-Exact Memory in Galahad Makes LLM Reading a One-Time Cost LLM Reading a One-Time Cost
CorbenicÂ
submitted a
paper to Daily Papers 2 days ago
CorbenicÂ
submitted a
paper to Daily Papers 10 days ago
CorbenicÂ
authored 2
papers 2 months ago
CorbenicÂ
submitted 2
papers to Daily Papers 3 months ago
CorbenicÂ
authored 2
papers 5 months ago
Merlin: Deterministic Byte-Exact Deduplication for Lossless Context Optimization in Large Language Model Inference
Paper • 2605.09990 • Published • 1
Byte-Exact Deduplication in Retrieval-Augmented Generation: A Three-Regime Empirical Analysis Across Public Benchmarks
Paper • 2605.09611 • Published • 1