Prosperity before Collapse: How Far Can Off-Policy RL Reach with Stale Data on LLMs? Paper • 2510.01161 • Published Oct 1, 2025 • 13
Multiverse: Your Language Models Secretly Decide How to Parallelize and Merge Generation Paper • 2506.09991 • Published Jun 11, 2025 • 55
HeadInfer: Memory-Efficient LLM Inference by Head-wise Offloading Paper • 2502.12574 • Published Feb 18, 2025 • 13