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

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

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

[NeurIPS25] Official repo for "Simplicity Prevails: Rethinking Negative Preference Optimization for LLM Unlearning"

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).

Usenix Security’25: Measuring Sample-level Unlearning Completeness

The project is for LLM unlearning, trustworthy AI.

The LLM Unlearning repository is an open-source project dedicated to the concept of unlearning in Large Language Models (LLMs). It aims to address concerns about data privacy and ethical AI by exploring and implementing unlearning techniques that allow models to forget unwanted or sensitive data. This ensures that AI models comply with privacy.

AAAI-25: On Effects of Steering Latent Representation for Large Language Model Unlearning

Improving LLM Unlearning Robustness via Random Perturbations

📝 Enable effective unlearning in LLMs with Partial Model Collapse, targeting specific information removal while maintaining overall model utility.