Tsung-Han (Hank) Lin

I am a Master's student in Computer Science at the NYU Courant Institute of Mathematical Sciences. My research interests lie in vision-language models and multimodal agents, with a focus on aligning multimodal systems to perceive and reason about the real world.

Before NYU, I was a research assistant with the UCLA NLP Group, advised by Prof. Kai-Wei Chang, where I worked on video agents, and with National Taiwan University's MSLab, advised by Prof. Shou-De Lin, where I worked on vision-language models for chip physical design in collaboration with the NVIDIA Design Automation Research Group. I hold a B.S. in Mathematics of Computation from UCLA.

Email  /  CV  /  LinkedIn  /  Google Scholar

News 🎉

• Sep 2026 — Started Master's at NYU Courant.
• Oct 2025 — Joined UCLA NLP Group.
• May 2025 — Multimodal Chip Physical Design Engineer Assistant accepted to ICLAD 2026!
• Sep 2024 — Joined NTU MSLab.
• May 2024 — VISUAL-ALPHASOCIAL accepted to ACL Findings 2025!
• Mar 2024 — Graduated from UCLA.

Publications

• 
  Multimodal Chip Physical Design Engineer Assistant
  Yun-Da Tsai, Chang-Yu Chao, Liang-Yeh Shen, Tsung-Han Lin, Haoyu Yang, Chia-Tung Ho, Yi-Chen Lu, Wen-Hao Liu, Shou-De Lin, Haoxing Ren
  ICLAD 2026 — paper
  
  A vision-language model framework for EDA workflows that predicts routing congestion and generates interpretable, actionable design suggestions to assist chip engineers.
• 
  VISUAL-ALPHASOCIAL: Benchmark and Self-Reflective Chain-of-Thought Generation for Visual Social Commonsense Reasoning
  Zongyu Lin, Zhikun Xu, Yixin Wan, Xingcheng Yao, Xiaohan Song, Tsung-Han Lin, Selina Song, Pranav Subbaraman, Kai-Wei Chang, Yizhou Sun
  ACL Findings 2025 — paper
  
  A benchmark for visual social commonsense reasoning that evaluates and elicits self-reflective chain-of-thought from vision-language models.

Template from Jon Barron.