Although large language models (LLMs) achieve effective safety alignment at the time of release, they still face various safety challenges. A key issue is that fine-tuning often compromises the safety alignment of LLMs. To address this issue, we propose a method named \textbf{IRR} (\textbf{I}dentify, \textbf{R}emove, and \textbf{R}ecalibrate for Safety Realignment) that performs safety realignment for LLMs. The core of IRR is to identify and remove unsafe delta parameters from the fine-tuned models, while recalibrating the retained ones. We evaluate the effectiveness of IRR across various datasets, including both full fine-tuning and LoRA methods. Our results demonstrate that IRR significantly enhances the safety performance of fine-tuned models on safety benchmarks, such as harmful queries and jailbreak attacks, while maintaining their performance on downstream tasks. The source code is available at: \url{https://anonymous.4open.science/r/IRR-BD4F}.
Separate the Wheat from the Chaff: A Post-Hoc Approach to Safety Re-Alignment for Fine-Tuned Language Models
A method called IRR enhances the safety of fine-tuned large language models by identifying and removing unsafe parameters and recalibrating the remaining ones, thereby improving performance on safety benchmarks without degrading downstream task performance.
- Year
- 2024
- Venue
- arXiv 2024
- Authors
- 4
- Hosting
- Abstract onlyARXIV-DEFAULT
Cite
Notes
Only stored in your browser.
Attribution
- Abstract & full text
- arxiv.org/abs/2412.11041ARXIV-DEFAULT
- TL;DR
- Semantic Scholar