Just Let Linear States Forget the Distant Past: Prefix Caching via Suffix Replay for Hybrid LLMs Paper • 2609.33477 • Published about 1 month ago • 9
Swift-SVD: Theoretical Optimality Meets Practical Efficiency in Low-Rank LLM Compression Paper • 2604.01609 • Published Apr 2 • 15
Pairwise RM: Perform Best-of-N Sampling with Knockout Tournament Paper • 2501.13007 • Published Jan 22, 2025 • 19