Data to Defense: The Role of Curation in Aligning Large Language Models Against Safety Compromise

ACL ARR 2025 May Submission3683 Authors

Abstract: Large language models (LLMs) are widely adapted for downstream applications through fine-tuning, a process named customization. However, recent studies have identified a vulnerability during this process, where malicious samples can compromise the robustness of LLMs and amplify harmful behaviors. To address this challenge, we propose an adaptive data curation approach allowing any text to be curated to enhance its effectiveness in counteracting harmful samples during customization. To avoid the need for additional defensive modules, we further introduce a comprehensive mitigation framework spanning the lifecycle of the customization process: before customization to immunize LLMs against future compromise attempts, during customization to neutralize risks, and after customization to restore compromised models. Experimental results demonstrate a significant reduction in compromising effects, achieving up to a 100% success rate in generating safe responses. By combining adaptive data curation with lifecycle-based mitigation strategies, this work represents a solid step forward in mitigating compromising risks and ensuring the secure adaptation of LLMs.

Paper Type: Long

Research Area: Ethics, Bias, and Fairness

Research Area Keywords: security/privacy

Contribution Types: Data analysis

Languages Studied: English

Keywords: Large Language Models, Robustness, AI Safety

Submission Number: 3683