[NeurIPS D&B '25] The one-stop repository for large language model (LLM) unlearning. Supports TOFU, MUSE, WMDP, and many unlearning methods with easy feature extensibility.

[ICLR24 (Spotlight)] "SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generation" by Chongyu Fan*, Jiancheng Liu*, Yihua Zhang, Eric Wong, Dennis Wei, Sijia Liu

The official implementation of ECCV'24 paper "To Generate or Not? Safety-Driven Unlearned Diffusion Models Are Still Easy To Generate Unsafe Images ... For Now". This work introduces one fast and effective attack method to evaluate the harmful-content generation ability of safety-driven unlearned diffusion models.

RWKU: Benchmarking Real-World Knowledge Unlearning for Large Language Models. NeurIPS 2024

[NeurIPS23 (Spotlight)] "Model Sparsity Can Simplify Machine Unlearning" by Jinghan Jia*, Jiancheng Liu*, Parikshit Ram, Yuguang Yao, Gaowen Liu, Yang Liu, Pranay Sharma, Sijia Liu

Continual Forgetting for Pre-trained Vision Models (CVPR 2024)

[ACL 2024] Code and data for "Machine Unlearning of Pre-trained Large Language Models"

[EMNLP 2024] To Forget or Not? Towards Practical Knowledge Unlearning for Large Language Models

[ACL 2025] Knowledge Unlearning for Large Language Models

[NeurIPS 2024] Large Language Model Unlearning via Embedding-Corrupted Prompts

Code for implementation of Unlearning Scanner Bias for MRI Harmonisation

Implementation of our unlearning method "Partial Model Collapse" introduced in the paper: "Model Collapse Is Not a Bug but a Feature in Machine Unlearning for LLMs" (Preprint).

Implementation of paper 'Reversing the Forget-Retain Objectives: An Efficient LLM Unlearning Framework from Logit Difference' [NeurIPS'24]

[ECCV24] "Challenging Forgets: Unveiling the Worst-Case Forget Sets in Machine Unlearning" by Chongyu Fan*, Jiancheng Liu*, Alfred Hero, Sijia Liu

Pytorch implementation of backdoor unlearning.

ERASURE: Redefining Privacy Through Selective Machine Unlearning

Implementation for MICCAI DART paper: 'Detecting Melanoma Fairly: Skin Tone Detection and Debiasing for Skin Lesion Classification'

Experiments for our CLEAR benchmark of unlearning methods in a multimodal setup