Xiaoyuan Liu

Scalable Supervising Software Agents with Patch Reasoner

The Pensieve Paradigm: Stateful Language Models Mastering Their Own Context

Castle-in-the-Air: Evaluating MLLM Visual Abilities on Human Cognitive Benchmarks

Curing "Miracle Steps'' in LLM Math Reasoning with Rubric Rewards

Trust, But Verify: A Self-Verification Approach to Reinforcement Learning with Verifiable Rewards

The First Few Tokens Are All You Need: An Efficient and Effective Unsupervised Prefix Fine-Tuning Method for Reasoning Models

Insight Over Sight? Exploring the Vision-Knowledge Conflicts in Multimodal LLMs