.png){kind=icon}
返回
Mitigating sensitive information leakage in LLMs4Code through machine unlearning
.png){kind=icon}
PRE
AI
DOI：10.1016/j.neunet.2026.108606.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
• Machine unlearning is a promising way to simultaneously mitigate the privacy concerns of LLMs4Code while maintaining their code generation capabilities at the same time. Specifically, unlearning can decrease the leak rate of AIXCoder by more than 50% while only bringing a negligible side effect to code generation. • After unlearning, LLMs4Code learn to adopt diverse forms to prevent the leakage of sensitivities, in which the most popular one is to replace the sensitive fields with variable names and abbreviations. • After unlearning, LLMs4Code become more likely to leak the privacy indirectly, which means they tend to leak the information that is not explicitly queried. This suggests that future works should also take into consideration the indirect privacy leakage for a more robust unlearning process. • All code and data in this study are publicly available at https://doi.org/10.5281/zenodo.14729266 .
期刊
IF:
6.3
论文数:
8.2K
被引数:
3.0W
机构
查看更多机构
引用论文
A Survey on Large Language Model (LLM) Security and Privacy: The Good, The Bad, and The Ugly关于大型语言模型 (LLM) 安全性和隐私的调查: 好，坏和丑陋
OAAI
.png){kind=icon}
OAAI分享收藏Your Code Secret Belongs to Me: Neural Code Completion Tools Can Memorize Hard-Coded Credentials
PREAI
分享收藏
分享收藏分享收藏 没有更多内容