Jie Lou

Scalable Oversight for Superhuman AI via Recursive Self-Critiquing

MemSearcher: Training LLMs to Reason, Search and Manage Memory via End-to-End Reinforcement Learning

Rethinking Reward Model Evaluation: Are We Barking up the Wrong Tree?

Critic-CoT: Boosting the reasoning abilities of large language model via Chain-of-Thought Critic

The Devil Is in the Details: Tackling Unimodal Spurious Correlations for Generalizable Multimodal Reward Models

On-Policy Fine-grained Knowledge Feedback for Hallucination Mitigation