π¨ Preprint out β Clinical Consensus SelectionοΌ
Inference-time report selection for radiology MLLMs.
About Me
π I’m a final-year PhD student at the University of Glasgow, working with the AI4BioMed Lab and the Information Retrieval Group (School of Computing Science).
π Founder of EvoScientist β harnessing vibe research to build self-evolving AI scientists that co-evolve with researchers and internalize scholarly taste and scientific judgment.
π€ I am also the formulator of the Three Laws of Self-Evolving AI Agents, a conceptual framework introduced in “A Comprehensive Survey of Self-Evolving AI Agents: A New Paradigm Bridging Foundation Models and Lifelong Agentic Systems”, for adaptive and collaborative AI Agents systems.
π¬ Open to collaborations and startup connections β feel free to reach out.
Interests
- Language Game
- Vibe Research
- AI Agents & AI4Health
- Multimodal Large Language Model
- Radiology Vision-language System
- Web3 & Blockchain
Education
PhD in Computing Science
University of Glasgow
Master of Information Technology & Project Management
University of Sydney
B.E. in Software Engineering
Dalian University of Technology
News
βοΈ Invited as Reviewer for NeurIPS 2026
Neural Information Processing Systems
π Paper Accepted β See You at ACL 2026!
We’re heading to San Diego for ACL 2026!
π€ Invited Talk at EvoAgentX Community!
Shared a talk about EvoScientist.
Selected Publications
Recent Publications
* indicate equal contributors.
(2026).
CCD: Mitigating Hallucinations in Radiology MLLMs via Clinical Contrastive Decoding.
ACL 2026.
(2026).
Automated Chest X-ray Report Generation Remains Unsolved.
PSB2026.
(2025).
RadEval: A framework for radiology text evaluation.
EMNLP 2025.