一作ACL 2026Poster· Natural Language Processing
Toward Secure Tuning: Mitigating Security Risks from Instruction Fine-Tuning
Yanrui Du · Fenglei Fan · Sendong Zhao · Jiawei Cao · Ming Ma · Danyang Zhao · Shuren Qi · Ting Liu · Bing Qin
摘要
Instruction Fine-Tuning (IFT) has emerged as a critical technique for customizing Large Language Models (LLMs) to meet diverse downstream applications. However, recent studies have revealed that IFT can compromise the built-in security mechanisms of LLMs, thereby posing significant security risks...