Resume
Academic Achievements
- Preprint 'EasyUUV: An LLM-Enhanced Universal and Lightweight Sim-to-Real Reinforcement Learning Framework for UUV Attitude Control': Proposes a lightweight, LLM-enhanced RL framework enabling zero-shot adaptation for UUV attitude control, validated on a low-cost 6DoF platform
- Preprint 'Ocean Diviner: A Diffusion-Augmented Reinforcement Learning for AUV Robust Control in the Underwater Tasks': Introduces a diffusion-augmented RL method to improve AUV control robustness in dynamic underwater environments
- IEEE Transactions on Mobile Computing (Major Revision) 'Never too Cocky to Cooperate: An FIM and RL-based USV-AUV Collaborative System for Underwater Tasks in Extreme Sea Conditions': Presents a USV–AUV collaborative system that significantly enhances underwater task performance under extreme sea conditions
- IEEE/RSJ IROS 2025 'Never too Prim to Swim: An LLM-Enhanced RL-based Adaptive S-Surface Controller for AUVs under Extreme Sea Conditions': Proposes an LLM-enhanced adaptive S-Surface controller for robust, multi-objective AUV control in extreme sea conditions
- IEEE Journal of Biomedical and Health Informatics 'Leveraging LLMs for Personalized Parkinson’s Disease Treatment': Explores using LLMs to support personalized treatment decision-making for Parkinson’s Disease
Background
- Second-year graduate student at Tsinghua University
- Research interests include multimodal large language models (LLMs), large-scale reinforcement learning, and their applications in embodied intelligence
- Focuses on underwater robotics, aiming to develop robust, adaptable, and generalizable control methods
- Applies reinforcement learning, learning from demonstrations (LfD), and multimodal LLMs to both terrestrial and underwater robotic systems
- Advocates for open-source and universally applicable research outcomes to benefit the broader research community