SDD: Self-Degraded Defense against Malicious Fine-tuning
ZiXuan Chen, Weikai Lu, Xin Lin, Ziqian Zeng
Abstract
Open-source Large Language Models (LLMs) often employ safety alignment methods to resist harmful instructions. However, recent research shows that maliciously fine-tuning these LLMs on harmful data can easily bypass these safeguards. To counter this, we theoretically uncover why malicious fine-tuning succeeds and identify potential defense strategies. Building on the theoretical analysis, we introduce the Self-Degraded Defense (SDD) framework. SDD encourages LLMs to produce high-quality but irrelevant responses to harmful prompts. When attackers attempt malicious fine-tuning, the general capability of the LLM aligned by SDD will significantly decrease, rendering it incapable of following harmful instructions. Our experimental results confirm SDD’s effectiveness against such attacks.Our code is available at https://github.com/ZeroNLP/SDD.
BibTeX
@inproceedings{chen-etal-2025-sdd,
title = "{SDD}: Self-Degraded Defense against Malicious Fine-tuning",
author = "Chen, ZiXuan and
Lu, Weikai and
Lin, Xin and
Zeng, Ziqian",
editor = "Che, Wanxiang and
Nabende, Joyce and
Shutova, Ekaterina and
Pilehvar, Mohammad Taher",
booktitle = "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = jul,
year = "2025",
address = "Vienna, Austria",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.acl-long.1412/",
doi = "10.18653/v1/2025.acl-long.1412",
pages = "29109--29125",
ISBN = "979-8-89176-251-0"
}