Wun Yu, Lucas Chan

MPhil in Computer Science at HKUST, AI Research

prof_pic.jpg

I rarely wear glassess.




I’m Lucas Chan, an MPhil student at HKUST, working under the guidance of Professor Yangqiu Song. My research spans embedding models and LLM reasoning. I hold a Bachelor’s degree with double major in Computer Science & Electronic Engineering (First Class Honours) from HKUST.

I have been working as an AI Research Intern at Amber Group, where my work focuses on LLM training and AI Agent applications. In my spare time, I also build some interesting projects. View my GitHub to explore those gripping projects.

Research Interests

  • Representation Learning Latent Space, Representation Similarity, and Manifold
  • Large Language Models Reasoning, training, alignment, and safety
  • AI Agents Autonomous systems, tool use, and multi-agent cooperation

News

Jan 30, 2026 My paper “Do Reasoning Models Enhance Embedding Models” is published on Arxiv! Feel free to check out the paper and the repository.
Jan 25, 2026 My paper “An Effective and Cost-Efficient Agentic Framework for Ethereum Smart Contract Auditing” is published on Arxiv, a work during the internship in Amber Group! Feel free to check out the paper.

Selected publications

  1. reasoning-embedding.png
    Do Reasoning Models Enhance Embedding Models?
    Wun Yu Chan, Shaojin Chen, Huihao Jing, and 5 more authors
    2026
  2. agentic-framework-radar.png
    An Effective and Cost-Efficient Agentic Framework for Ethereum Smart Contract Auditing
    Xiaohui Hu*, Wun Yu Chan*, Yuejie Shi, and 5 more authors
    2026