This repository contains a collection of papers on LLM-Unlearning.

• An introduction to Unlearning: Ken Liu blog

• Rethinking Machine Unlearning for Large Language Models: paper

• TOFU - A Task of Fictitious Unlearning for LLMs: paper

• WMDP Benchmark: paper

• Who’s harry potter? approximate unlearning in LLMs: paper

• Knowledge sanitization of large language models: paper

• TOFU - A Task of Fictitious Unlearning for LLMs: paper

• Can we edit the factual knowledge in LLMs: paper

• In-context unlearning- Language models as few shot unlearners: paper

• Gradient Ascent vs Gradient Descent: blog

• Knowledge unlearning for mitigating privacy risks in language models Knowledge unlearning for mitigating privacy risks in language models: paper

• Large language model unlearning: paper

• Can we edit the factual knowledge in LLMs: paper

• Detecting and editing privacy neurons in pre-trained language models: paper

• Can sensitive information be deleted from LLMs? objectives for defending against extraction attacks: paper

• Privacy adhering machine un-learning in NLP: paper

• Unlearning bias in language models by partitioning gradients: paper

• Controllable text generation with reinforced unlearning: paper

• A general machine unlearning framework based on knowledge gap alignment: paper

• Unlearn what you want to forget: Efficient unlearning for LLMs: paper

• Editing models with task arithmetic: paper

• Composing parameter-efficient modules with arithmetic operations: paper