Hi! I am a second-year CS PhD student jointly at NYU Courant and NYU Shanghai, mentored by Professor Chen Zhao. I also closely work with Professor Xi Ye at University of Alberta. Previously, I received my B.S. degree in Data Science (with Honors) from New York University Shanghai, where I was advised by Professor Wilson Tam and Professor Mathieu Lauriere. My primary research interests are in Natural Language Processing and Deep Learning. Here are some topics that I have been thinking about lately:

1. How to best empower LLMs with the information retrieval ability, so that they can reliably solve complex real-world agentic tasks (e.g. Deep Research report generation) that involves knowledge-intensive reasoning? How can (Deep) Research Agents best collaborate and interact with human researchers?
2. Can we automate the error characterization process of agentic traces, or long-form generation in general? Given a set of generations, can LLM accurately induce and summarize what the error patterns are?
3. I am always fascinated by novel and smart decoding strategies and embedding learning methods!

If you have any thoughts on these, shoot me an email and I am more than happy to chat with you!

News

• [July 2026] Mr Dre is awarded SAC Highlight and selected as Best Paper candidates at ACL 2026! Check out our paper and code!
• [May 2026] I am returning to AWS as an Applied Science intern in New York, working on long-horizon agent efficiency.
• [February 2026] Check out LLaDA 2.1, large-scale diffusion language models with token editing abilities! Glad to have contributed to the on-policy SFT and editing-specific training strategies.
• [December 2025] Joined Ant Research Center as a Research Intern to work on LLaDA post-training.
• [May 2025] I joined AWS Bedrock as an Applied Science intern, working on explainability of agents.
• [May 2025] Our paper on multi-answer QA is accepted to ACL 2025 Findings. Check it out here!
• [September 2024] Starting my PhD at NYU Courant and NYU Shanghai!

(Last Update: July 2026)