Pruning for Protection: Increasing Jailbreak Resistance in Aligned LLMs Without Fine-Tuning

This paper investigates the impact of model compression on the way Large Language Models (LLMs) process prompts, particularly concerning jailbreak resistance.We show that moderate WANDA pruning (Sun et al., 2023) can enhance resistance to jailbreaking attacks without fine-tuning, while maintaining performance on standard benchmarks.To systematically evaluate this safety enhancement, we introduce a dataset of 225 harmful tasks across five categories.Our analysis of LLaMA-2 Chat (Touvron et al., 2023), Vicuna 1.3 (Chiang et al., 2023), and Mistral Instruct v0.2 (Jiang et al., 2023) reveals that pruning benefits correlate with initial model safety levels.We interpret these results by examining changes in attention patterns and perplexity shifts, demonstrating that pruned models exhibit sharper attention and increased sensitivity to artificial jailbreak constructs.We extend our evaluation to the Ad-vBench harmful behavior tasks and the GCG attack method (Zou et al., 2023).We find that LLaMA-2 is much safer on AdvBench prompts than on our dataset when evaluated with manual jailbreak attempts, and that pruning is effective against both automated attacks and manual jailbreaking on Advbench.