Chongyu Fan

Chongyu Fan

Ph.D student @ Michigan State University

👋 Welcome to Chongyu’s home page!

🔹 I am Chongyu Fan (樊翀宇), a second-year Ph.D. student in Computer Science at Michigan State University, advised by Prof. Sijia Liu. I received my B.Eng. from Huazhong University of Science and Technology in 2024.

🔹 I am currently a Student Researcher at ByteDance Seed, working on multimodal foundation models.

🔹 My research interests include:

  • Reinforcement learning for agents and foundation models
  • Efficient reasoning and test-time scaling
  • Machine unlearning and alignment

🔹 Feel free to reach out by email if you’re interested in collaborating.

🔥 News

2025.10: ✨ One first-author paper LLM Unlearning Bench has been made public on arXiv!
2025.09: 🎉 One first-author paper SimNPO has been accepted in NeurIPS 2025!
2024.09: 🎉 One paper ABO has been accepted in NeurIPS 2025 as spotlight!
2025.08: 🎉 One first-author paper Reasoning Unlearn has been accepted in EMNLP 2025 Main!
2025.06: ✨ One first-author paper CyclicReflex has been made public on arXiv!
2025.06: ✨ One paper EPiC has been made public on arXiv!
2025.05: 🎉 One first-author paper Smooth Unlearn has been accepted in ICML 2025!
2024.09: 🎉 Two papers UnlearnCanvas and AdvUnlearn have been accepted in NeurIPS 2024!
2024.07: 🎉 One first-author paper Challenging Forgets has been accepted in ECCV 2024!
2024.01: 🎉 One first-author paper SalUn has been accepted in ICLR 2024 as spotlight!

🎯 First-Authored Publications

Simplicity Prevails: Rethinking Negative Preference Optimization for LLM Unlearning

Chongyu Fan*, Jiancheng Liu*, Licong Lin*, Jinghan Jia, Ruiqi Zhang, Song Mei, Sijia Liu

NeurIPS 2025

[Paper] [Code]

Reasoning Model Unlearning: Forgetting Traces, Not Just Answers, While Preserving Reasoning Skills

Changsheng Wang*, Chongyu Fan*, Yihua Zhang, Jinghan Jia, Dennis Wei, Parikshit Ram, Nathalie Baracaldo, Sijia Liu

EMNLP 2025 Main

[Paper] [Code]

Towards LLM Unlearning Resilient to Relearning Attacks: A Sharpness-Aware Minimization Perspective and Beyond

Chongyu Fan*, Jinghan Jia*, Yihua Zhang, Anil Ramakrishna, Mingyi Hong, Sijia Liu

ICML 2025

[Paper] [Code]

Challenging Forgets: Unveiling the Worst-case Forget Sets in Machine Unlearning

Chongyu Fan*, Jiancheng Liu*, Alfred Hero, Sijia Liu

ECCV 2024

[Paper] [Code]

SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generation

Chongyu Fan*, Jiancheng Liu*, Yihua Zhang, Dennis Wei, Eric Wong, Sijia Liu

ICLR 2024 Spotlight

[Paper] [Code] [Website]

📄 CV (Updated: 2026-01-03)

📄

English CV

Here is my CV in English.

📞 Contact

📧

Email

fanchon2@msu.edu

chongyu.fan93@gmail.com

📈 Experience

🎓 Education

  • 2024.09–Present, Ph.D. in CS, Michigan State University
  • 2020.09–2024.06, B.E. in Robotics (Outstanding Graduate), Huazhong University of Science and Technology

💻 Work

  • 2025.12–Present, Student Researcher, ByteDance Seed
  • 2025.05–2025.09, Research Scientist Intern, TikTok

🚚 Services

Program Committee/Reviewer

Conferences

  • The International Conference on Machine Learning (ICML): ICML25, ICML26
  • The International Conference on Learning Representations (ICLR): ICLR25, ICLR26
  • The Annual Conference on Neural Information Processing Systems (NeurIPS): NeurIPS25