[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.

A Comprehensive Survey of Forgetting in Deep Learning Beyond Continual Learning. TPAMI, 2024.

Testing membership inference attacks on Deep learning models (LSTM, CNN);

Loss and Likelihood Based Membership Inference of Diffusion Models

A unified evaluation suite for membership inference and machine text detection.

apcMIA: Adaptive Perturbation-assisted Contrastive Membership Inference Attack This repository contains the implementation of APCMIA, a black-box membership inference attack framework that combines adaptive perturbation with contrastive learning. It is designed to uncover subtle membership signals in well-generalized.

Revisiting the LiRA Membership Inference Attack Under Realistic Assumptions