ljcc0930 / tofu

Landing Page for TOFU

Hi there 👋

My name is Jiancheng (JC) Liu. I am a graduate student at Michigan State University, currently work at OPTML Group. My recent research focuses on scalable and trustworthy AI.

• I am planning to enroll as a doctoral student in 2025. Should you have any promising opportunities, please feel free to contact me!

My recent works about machine unlearning (MU)

Paper Title	Venue	Repository
Defensive Unlearning with Adversarial Training for Robust Concept Erasure in Diffusion Models	In Submission
SOUL: Unlocking the Power of Second-Order Optimization for LLM Unlearning	In Submission
UnlearnCanvas: A Stylized Image Dataset to Benchmark Machine Unlearning for Diffusion Models	In Submission
Challenging Forgets: Unveiling the Worst-Case Forget Sets in Machine Unlearning	ECCV'24
To Generate or Not? Safety-Driven Unlearned Diffusion Models Are Still Easy To Generate Unsafe Images... For Now	ECCV'24
SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generation	ICLR'24 (Spotlight)
Model Sparsity Can Simplify Machine Unlearning	NeurIPS'23 (Spotlight)