The Janus Interface: How Fine-Tuning in Large Language Models Amplifies the Privacy Risks

​In this session, We are welcoming Xiaoyi Chen from Indiana University who co-authored the paper "The Janus Interface: How Fine-Tuning in Large Language Models Amplifies the Privacy Risks".

The paper can be found here

The research introduces a novel attack, Janus, which exploits the fine-tuning interface to recover forgotten PIIs from the pre-training data in LLMs. See you there!