Popular repositories Loading

1. GradAttack GradAttack Public

   GradAttack is a Python library for easy evaluation of privacy risks in public gradients in Federated Learning, as well as corresponding mitigation strategies.

   Python 185 38

2. Jailbreak_LLM Jailbreak_LLM Public

   Jupyter Notebook 157 14

3. FILM FILM Public

   Official repo for the paper: Recovering Private Text in Federated Learning of Language Models (in NeurIPS 2022)

   Python 57 7

4. kNNLM_privacy kNNLM_privacy Public

   Official implementation of Privacy Implications of Retrieval-Based Language Models (EMNLP 2023). https://arxiv.org/abs/2305.14888

   Python 36 3

5. jailbreak-llm jailbreak-llm Public

   JavaScript