Max W.F. Ku

I’m a first-year PhD student in Computer Science at the University of Waterloo, Faculty of Mathematics, where I’m fortunate to be advised by Prof. Wenhu Chen. I am currently interning as a Research Scientist at NVIDIA, under Ming-Yu Liu.

My research lies at the intersection of generative AI, visual content creation, and model interpretability. At the heart of my work is a simple but ambitious goal:

To make generative visuals fully controllable across science, communication, and creative applications.

While visuals remain my core focus, I’m increasingly curious about how they can integrate with physical reasoning and scientific understanding. I believe controllability in generative models should go beyond aesthetics, extending to physical coherence and alignment with how we perceive the world.

My work spans

Controllable Editing and Generation (I prioritize editing over generation)
Multimodal Agentic Systems (Visuals + X)
Interpretability and Explainable AI
Creative Applications in Entertainment, Education, and Science

Professional Activities

Reviewed for: ICLR, NeurIPS, ICML, SIGGRAPH Asia, SIGGRAPH, TVCG, ACL, EMNLP

Community

I lead GGG, a community-driven group dedicated to sharing and discussing papers on Generative AI.
I host Pro-bono Office Hour to share advice with students from underrepresented backgrounds.

Misc

My Blog, where I keep my reading notes and various logs.
I was a member of the HK PolyU Robotics Team during ABU Robocon 2019-2021. A playlist.
I used to compete in official gaming tournament UGC League Team Fortress 2 Highlander.
“Wing Fung” (with the space) is my first name and “Ku” is my last name. “Max” is the commonly used “english name” that is not part of my legal name. This is common in Hong Kong.

news

Jun 15, 2025	Achieved a total of 1000 citations.
Jun 11, 2025	DisProtEdit got accepted to 2025 ICML GenBio workshop and FM4LS workshop.
Jun 02, 2025	Joined NVIDIA Deep Imagination Research as an intern for Summer 2025.
May 15, 2025	TheoremExplainAgent got accepted to ACL 2025 Main (Oral)!
Nov 03, 2024	AnyV2V got accepted to TMLR 2024!

selected publications

ACL 2025 Oral

TheoremExplainAgent: Towards Multimodal Explanations for LLM Theorem Understanding

Max Ku*, Thomas Chong*, Jonathan Leung, Krish Shah, Alvin Yu, and 1 more author

In The 63rd Annual Meeting of the Association for Computational Linguistics , 2025

arXiv Bib Code Website

@inproceedings{ku2025theoremexplainagentmultimodalexplanationsllm,
  title = {TheoremExplainAgent: Towards Multimodal Explanations for LLM Theorem Understanding},
  booktitle = {The 63rd Annual Meeting of the Association for Computational Linguistics},
  author = {Ku*, Max and Chong*, Thomas and Leung, Jonathan and Shah, Krish and Yu, Alvin and Chen, Wenhu},
  year = {2025},
  selected = true,
  bibtex_show = false
}

NeurIPS 2024

GenAI Arena: An Open Evaluation Platform for Generative Models

Dongfu Jiang*, Max Ku*, Tianle Li*, Yuansheng Ni, Shizhuo Sun, and 2 more authors

In The Conference on Neural Information Processing Systems , 2024

arXiv Bib Website

@inproceedings{jiang2024genai,
  title = {GenAI Arena: An Open Evaluation Platform for Generative Models},
  author = {Jiang*, Dongfu and Ku*, Max and Li*, Tianle and Ni, Yuansheng and Sun, Shizhuo and Fan, Rongqi and Chen, Wenhu},
  year = {2024},
  booktitle = {The Conference on Neural Information Processing Systems},
  selected = true,
  bibtex_show = false
}

TMLR 2024

AnyV2V: A Tuning-Free Framework For Any Video-to-Video Editing Tasks

Max Ku*, Cong Wei*, Weiming Ren*, Harry Yang, and Wenhu Chen

Transactions on Machine Learning Research, 2024

arXiv Bib Code Website

@article{ku2024anyv2v,
  title = {AnyV2V: A Tuning-Free Framework For Any Video-to-Video Editing Tasks},
  author = {Ku*, Max and Wei*, Cong and Ren*, Weiming and Yang, Harry and Chen, Wenhu},
  journal = {Transactions on Machine Learning Research},
  year = {2024},
  selected = true,
  bibtex_show = false
}

ACL 2024

VIEScore: Towards Explainable Metrics for Conditional Image Synthesis Evaluation

Max Ku, Dongfu Jiang, Cong Wei, Xiang Yue, and Wenhu Chen

In The 62nd Annual Meeting of the Association for Computational Linguistics , 2024

arXiv Bib Code Website

@inproceedings{Ku2023VIEScoreTE,
  title = {VIEScore: Towards Explainable Metrics for Conditional Image Synthesis Evaluation},
  author = {Ku, Max and Jiang, Dongfu and Wei, Cong and Yue, Xiang and Chen, Wenhu},
  booktitle = {The 62nd Annual Meeting of the Association for Computational Linguistics},
  year = {2024},
  selected = true,
  bibtex_show = false
}

ICLR 2024

ImagenHub: Standardizing the evaluation of conditional image generation models

Max Ku, Tianle Li, Kai Zhang, Yujie Lu, Xingyu Fu, and 2 more authors

In The 12th International Conference on Learning Representations , 2024

arXiv Bib Code Website

@inproceedings{ku2024imagenhub,
  title = {ImagenHub: Standardizing the evaluation of conditional image generation models},
  author = {Ku, Max and Li, Tianle and Zhang, Kai and Lu, Yujie and Fu, Xingyu and Zhuang, Wenwen and Chen, Wenhu},
  booktitle = {The 12th International Conference on Learning Representations},
  year = {2024},
  twitter = {https://twitter.com/vinesmsuic/status/1717564355212951701},
  selected = true,
  bibtex_show = false
}