About
I am a second-year Ph.D. student at the College of Computing and Data Science, Nanyang Technological University, advised by Prof. Alvin Chan Guo Wei. My research explores new ways to combine large language models with domain foundation models, aiming for integrations that are both more efficient and more effective. The current focus is on scientific domains — small molecules and proteins — where structured FM representations can complement LLM reasoning.
Before NTU, I received my B.Eng. in Computer Science (with a joint degree in Smart Finance & Blockchain Finance) from the University of Electronic Science and Technology of China (UESTC). I have been fortunate to work with researchers at the Shanghai AI Lab and the NUS HPC-AI Lab on evaluation of generative models and on graph dataset distillation.
I am open to all kinds of novel research directions and always excited by unexplored problems. If you have a collaboration in mind — across disciplines or methodologies — feel free to drop me an email to discuss.
News
- May 2026 Recognized as an ICML 2026 Gold Reviewer (top reviewers based on area-chair ratings).
- Dec 2025 Our paper on strategic multimodal representation alignment is accepted to AAAI 2026.
- Sep 2025 Our paper on training-free multi-modal reasoning via In-Context Representation Learning is accepted to NeurIPS 2025.
- Jan 2025 Started my Ph.D. at NTU, advised by Prof. Alvin Chan.
- Sep 2024 Awarded the NTU Research Scholarship for Ph.D. studies.
- Sep 2024 Two papers accepted to NeurIPS 2024 — one first-author (T2V evaluation), one Spotlight (ConvBench).
- May 2024 Our paper on lossless graph dataset condensation is accepted to ICML 2024.
Publications
Bold = me. * equal contribution. † corresponding author. Full list on Google Scholar.
Can LLMs Reason Over Non-Text Modalities in a Training-Free Manner? A Case Study with In-Context Representation Learning
Rethinking Human Evaluation Protocol for Text-to-Video Models: Enhancing Reliability, Reproducibility, and Practicality
ConvBench: A Multi-Turn Conversation Evaluation Benchmark with Hierarchical Capability for Large Vision-Language Models
Navigating Complexity: Toward Lossless Graph Condensation via Expanding Window Matching
ADGCN: A Weakly Supervised Framework for Anomaly Detection in Social Networks
Phase Detection Methods for Alzheimer's Disease Based on Improved Convolutional Neural Network Fine-tuning
Fast Classification of Alzheimer's Disease Based on Improved Convolutional Neural Network
Awards & Scholarships
- ICML 2026 Gold Reviewer
- NTU Research Scholarship (R-RSS)
- Excellent First-class Scholarship
- Xiaomi Scholarship
- Tencent First-class Scholarship
- NTU Project Scholarship
- Special Prize · 7th UESTC "Internet Plus" Innovation & Entrepreneurship Competition
Academic Service
Conference Reviewer
- ICLR — 2026
- NeurIPS — 2025, 2026
- ICML — 2025, 2026 (Gold Reviewer)
- ICLR Workshop on AI for Nucleic Acids (AI4NA) — 2025
Workshop Committee
- ECCV 2024 Workshop on Dataset Distillation Challenge — Committee Member