Xin Wang
AI Researcher
Oxford, United Kingdom
My interests mainly lie in large language model (LLM) post-training, adaptation, reasoning, and multimodal learning, with applications in healthcare, biomedicine, and social computing. My long-term goal is to develop AI systems that are not only more capable, but also more understandable and better aligned with human needs and values.
For the most recent publications, please visit my Google Scholar page.
I am currently a Researcher at The Alan Turing Institute and a Visiting Researcher at the University of Oxford, where I previously worked as a postdoctoral researcher with Prof. David Clifton. I received my Ph.D. in Computer Science and Technology from Tsinghua University under the supervision of Prof. Ling Feng.
Outside of research, I enjoy climbing and travelling — I hope to explore as much of the world as I can!
news
| Jun 2026 | Released Mental-R1, which proposes an RL algorithm to align LLM reasoning with human cognitive processes. Available on arXiv and GitHub. |
|---|---|
| Sep 2025 | Our paper on optimization-inspired few-shot adaptation of LLMs was accepted to NeurIPS 2025 as a Spotlight. |
| May 2025 | Gave invited talks at the Web Conference 2025 Health Day and the University of Sydney. |
| Jan 2025 | MISE, my first-author work on avoiding catastrophic forgetting for stressor estimation, was accepted to WWW 2025 as an Oral. |
