Zuxin Liu

Webscale-RL: Automated Data Pipeline for Scaling RL Data to Pretraining Levels

Test-Time Adaptation for LLM Agents via Environment Interaction

UserRL: Training Interactive User-Centric Agent via Reinforcement Learning

ToolLibGen: Scalable Automatic Tool Creation and Aggregation for LLM Reasoning

LoCoBench: A Benchmark for Long-Context Large Language Models in Complex Software Engineering

Behavior Injection: Preparing Language Models for Reinforcement Learning

Diversity Empowers Intelligence: Integrating Expertise of Software Engineering Agents

LAM Simulator: Advancing Large Action Model Training for Agent via Online Exploration and Feedback Simulation

MobileAIBench: Benchmarking LLMs and LMMs for On-Device Use Cases

Language Models are Hidden Reasoners: Unlocking Latent Reasoning Capabilities via Self-Rewarding