Sentence Embedding Leaks More Information than You Expect: Generative Embedding Inversion Attack to Recover the Whole Sentence

1 code implementation • 4 May 2023 • Haoran Li, Mingshi Xu, Yangqiu Song

In this work, we further investigate the information leakage issue and propose a generative embedding inversion attack (GEIA) that aims to reconstruct input sequences based only on their sentence embeddings.

Language Modelling Sentence +2

Multi-step Jailbreaking Privacy Attacks on ChatGPT

1 code implementation • 11 Apr 2023 • Haoran Li, Dadi Guo, Wei Fan, Mingshi Xu, Jie Huang, Fanpu Meng, Yangqiu Song

With the rapid progress of large language models (LLMs), many downstream NLP tasks can be well solved given appropriate prompts.