Language Models are Homer Simpson! Safety Re-Alignment of Fine-tuned Language Models through Task Arithmetic
Rishabh Bhardwaj, Duc Anh, Soujanya Poria
Abstract
Aligned language models face a significant limitation as their fine-tuning often results in compromised safety.To tackle this, we propose a simple method RESTA that performs LLM safety realignment.RESTA stands for REstoring Safety through Task Arithmetic.At its core, it involves a simple arithmetic addition of a safety vector to the weights of the compromised model.We demonstrate the effectiveness of RESTA in both parameter-efficient and full fine-tuning, covering a wide range of downstream tasks, including instruction following in Chinese, English, and Hindi, as well as problem-solving capabilities in Code and Math.We also showcase the generalizability of RESTA on three existing safety evaluation benchmarks and a multilingual benchmark dataset proposed as a part of this work, consisting of 550 harmful questions covering 11 categories, each with 5 sub-categories of harm.Overall, RESTA decreases the harmfulness of the compromised model from 18.6% to 5.1% and from 9.2% to 1.5% in parameter-efficient and full finetuning, respectively, while maintaining most of the model's performance on the task.We release the source code at: https://github. com/declare-lab/resta.