Yuhua Jiang

Risk-Sensitive Reinforcement Learning for Alleviating Exploration Dilemmas in Large Language Models

PAG: Multi-Turn Reinforced LLM Self-Correction with Policy as Generative Verifier

OPRIDE: Efficient Offline Preference-based Reinforcement Learning via In-Dataset Exploration

GlobeDiff: State Diffusion Process for Partial Observability in Multi-Agent System

Episodic Novelty Through Temporal Distance

Fewer May Be Better: Enhancing Offline Reinforcement Learning with Reduced Dataset

Maximum Next-State Entropy for Efficient Reinforcement Learning

Fewer Questions, Better Answers: Efficient Offline Preference-based Reinforcement Learning via In-Dataset Exploration