Resume
Academic Achievements
- Multiple papers accepted at ICML 2025, including 'BRiTE' (equal contribution) and 'Reward-Augmented Data Enhances Direct Preference Alignment of LLMs' (equal contribution)
- Published 'Provably Mitigating Overoptimization in RLHF' at NeurIPS 2024
- Published 'Self-Exploring Language Models' in Transactions on Machine Learning Research (TMLR), awarded Best Paper at ICML AutoRL Workshop 2024
- Published 'Reason for Future, Act for Now' (equal contribution) and 'Adaptive-Gradient Policy Optimization' at ICML 2024
- Published 'Model-Based Reparameterization Policy Gradient Methods' at NeurIPS 2023 (co-advised by Prof. Zhaoran Wang and Prof. Tuo Zhao)
- Published 'Adaptive Barrier Smoothing for First-Order Policy Gradient with Contact Dynamics' at ICML 2023
- Multiple 2025 preprints on LLM reasoning and RL, e.g., 'Learning to Reason as Action Abstractions' and 'Beyond Markovian'
- Work featured by MIT Tech Review China and HF Daily Papers