Fan Lai

OPPO: Accelerating PPO-based RLHF via Pipeline Overlap

CORRECT: COndensed eRror RECognition via knowledge Transfer in multi-agent systems

XRPO: Pushing the Limits of GRPO with Targeted Exploration and Exploitation

Inv-Entropy: A Fully Probabilistic Framework for Uncertainty Quantification in Language Models

Act Only When It Pays: Efficient Reinforcement Learning for LLM Reasoning via Selective Rollouts

HyGen: Efficient LLM Serving via Elastic Online-Offline Request Co-location