Scholar
Xiangyan Liu
Google Scholar ID: lsFTg-0AAAAJ
National University of Singapore
Citations & Impact
Citations
161
H-index
6
i10-index
6
Publications
8
Co-authors
0
Publications
Resume (English only)
Academic Achievements
- MCPMark: A Benchmark for Stress-Testing Realistic and Comprehensive MCP Use (Preprint, 2025)
- NoisyRollout: Reinforcing Visual Reasoning with Data Augmentation (NeurIPS 2025)
- CodexGraph: Bridging Large Language Models and Code Repositories via Code Graph Databases (NAACL 2025)
- Towards Robust Multi-Modal Reasoning via Model Selection (ICLR 2024)
- Online Distillation-Enhanced Multi-Modal Transformer for Sequential Recommendation (MM 2023, Oral)
Research Experience
- Before starting his PhD, he gained valuable industry experience through internships at Shanghai AI Lab and Alibaba Tongyi. He also worked with Prof. Tao Lin at Westlake University, whose mentorship deeply inspired him. Recently, he has been working closely with Dr. Tianyu Pang, Dr. Chao Du, and Dr. Longxu Dou at Sea AI Lab.
Education
- Second-year PhD student in Computer Science at the National University of Singapore, advised by Prof. Michael Qizhe Shieh.
Background
- Research interests include multimodal and agentic foundation models, focusing on solutions that are simple, generalizable, and scalable.
Miscellany
- Feel free to reach out if you would like to chat or discuss any ideas!
Co-authors