I am a Ph.D. Candidate in Artificial Intelligence at The Hong Kong University of Science and Technology (Guangzhou), advised by Prof. Hui Xiong. I also work closely with Prof. Xuming Hu, Dr. Aiwei Liu at WeChat AI, and Dr. Weiyu Guo at MMLab, CUHK. I am currently a Research Intern at Tencent, working on generative retrieval.
Research interests. My research studies how language models acquire and leverage information to adapt and make decisions, with a focus on test-time self-evolution, generative retrieval, and retrieval-augmented generation.
I am open to industry research opportunities in LLM post-training and related areas.
News
- One survey paper on unsupervised post-training was accepted by EMNLP 2026.
- One paper on political ideology detection and prediction was accepted by ACM TIST.
- Started working as a Research Intern at Tencent.
- SafeBuild-Bench was accepted by the KDD 2026 Datasets & Benchmarks Track.
- One paper on VLM unlearning was accepted by CVPR 2026.
- One paper on source-free domain adaptation was accepted by AAAI 2026.
- Two papers on key frame search for long-video understanding and spatial prompting for VLMs were accepted by NeurIPS 2025.
Selected Publications
Full list on Google Scholar†: Co-first Authors; ‡: Co-second Authors; ∗: Corresponding Authors.
For research discussions or industry opportunities, the fastest way to reach me is email.