Wenlin Yao

Interpreting and Steering LLMs with Mutual Information-based Explanations on Sparse Autoencoders

Soundness-Aware Level: A Microscopic Signature that Predicts LLM Reasoning Potential

Improving Sampling Efficiency in RLVR through Adaptive Rollout and Response Reuse

Ask a Strong LLM Judge when Your Reward Model is Uncertain

DSBench: How Far Are Data Science Agents from Becoming Data Science Experts?

DOTS: Learning to Reason Dynamically in LLMs via Optimal Reasoning Trajectories Search

DeFine: Enhancing LLM Decision-Making with Factor Profiles and Analogical Reasoning

Interpreting and Steering LLM Representations with Mutual Information-based Explanations on Sparse Autoencoders

SePPO: Semi-Policy Preference Optimization for Diffusion Alignment

HDFlow: Enhancing LLM Complex Problem-Solving with Hybrid Thinking and Dynamic Workflows

MultiMedia-Agent: A Multimodal Agent for Multimedia Content Generation