Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2603.11388 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914386856116224 |
|---|---|
| author | Xue, Zhiyu Qi, Zimo Liu, Guangliang Chen, Bocheng Pedarsani, Ramtin |
| author_facet | Xue, Zhiyu Qi, Zimo Liu, Guangliang Chen, Bocheng Pedarsani, Ramtin |
| contents | Safety alignment aims to ensure that large language models (LLMs) refuse harmful requests by post-training on harmful queries paired with refusal answers. Although safety alignment is widely adopted in industry, the overrefusal problem where aligned LLMs also reject benign queries after safety alignment post-training, remains insufficiently studied. Such an issue degrades the usability of safety alignment in real-world applications. In this paper, we examine how overrefusal arises under safety alignment, and propose a mitigation strategy inspired by our findings. We define refusal triggers as linguistic cues in the training data that elicit refusal responses, safety alignment encourages LLMs to associate refusal triggers within a training sample with refusal responses, leading aligned LLMs to refuse harmful queries. However, the refusal triggers include not only harmful linguistic cues but also non-harmful cues, therefore causing overrefusal to benign queries. Building on this mechanistic analysis, we propose a method that explicitly considers refusal triggers in the safety alignment fine-tuning. Empirical results demonstrate that our approach achieves a more favorable trade-off between defense against jailbreak attacks and responsiveness to benign queries, outperforming prior methods. Warning: this paper contains harmful and biased sentences. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_11388 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Deactivating Refusal Triggers: Understanding and Mitigating Overrefusal in Safety Alignment Xue, Zhiyu Qi, Zimo Liu, Guangliang Chen, Bocheng Pedarsani, Ramtin Artificial Intelligence Safety alignment aims to ensure that large language models (LLMs) refuse harmful requests by post-training on harmful queries paired with refusal answers. Although safety alignment is widely adopted in industry, the overrefusal problem where aligned LLMs also reject benign queries after safety alignment post-training, remains insufficiently studied. Such an issue degrades the usability of safety alignment in real-world applications. In this paper, we examine how overrefusal arises under safety alignment, and propose a mitigation strategy inspired by our findings. We define refusal triggers as linguistic cues in the training data that elicit refusal responses, safety alignment encourages LLMs to associate refusal triggers within a training sample with refusal responses, leading aligned LLMs to refuse harmful queries. However, the refusal triggers include not only harmful linguistic cues but also non-harmful cues, therefore causing overrefusal to benign queries. Building on this mechanistic analysis, we propose a method that explicitly considers refusal triggers in the safety alignment fine-tuning. Empirical results demonstrate that our approach achieves a more favorable trade-off between defense against jailbreak attacks and responsiveness to benign queries, outperforming prior methods. Warning: this paper contains harmful and biased sentences. |
| title | Deactivating Refusal Triggers: Understanding and Mitigating Overrefusal in Safety Alignment |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2603.11388 |