XingYu

DeepEyes: Incentivizing "Thinking with Images" via Reinforcement Learning

Scalable Oversight for Superhuman AI via Recursive Self-Critiquing

DeepEyesV2: Toward Agentic Multimodal Model

Towards Agentic Self-Learning LLMs in Search Environment

Think When You Need: Self-Adaptive Chain-of-Thought Learning

Refusal Falls off a Cliff: How Safety Alignment Fails in Reasoning?

Probabilistic Uncertain Reward Model

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