Zichen Liu

Variational Reasoning for Language Models

GEM: A Gym for Generalist LLMs

SPIRAL: Self-Play on Zero-Sum Games Incentivizes Reasoning via Multi-Agent Multi-Turn Reinforcement Learning

Language Models Can Learn from Verbal Feedback Without Scalar Rewards

Sample-Efficient Alignment for LLMs

Reinforcing General Reasoning Without Verifiers

SynthRL: Scaling Visual Reasoning with Verifiable Data Synthesis

Understanding R1-Zero-Like Training: A Critical Perspective

Continual Reinforcement Learning by Planning with Online World Models

Efficient Process Reward Model Training via Active Learning

Optimizing Anytime Reasoning via Budget Relative Policy Optimization

Bootstrapping Language Models with DPO Implicit Rewards

Sample-Efficient Alignment for LLMs

Sample Efficient Alignment for LLMs