SafeBehavior: Simulating Human-Like Multistage Reasoning to Mitigate Jailbreak Attacks in Large Language Models
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911185923735552 |
|---|---|
| author | Zhao, Qinjian Wang, Jiaqi Gao, Zhiqiang Dou, Zhihao Abuhaija, Belal Huang, Kaizhu |
| author_facet | Zhao, Qinjian Wang, Jiaqi Gao, Zhiqiang Dou, Zhihao Abuhaija, Belal Huang, Kaizhu |
| contents | Large Language Models (LLMs) have achieved impressive performance across diverse natural language processing tasks, but their growing power also amplifies potential risks such as jailbreak attacks that circumvent built-in safety mechanisms. Existing defenses including input paraphrasing, multi step evaluation, and safety expert models often suffer from high computational costs, limited generalization, or rigid workflows that fail to detect subtle malicious intent embedded in complex contexts. Inspired by cognitive science findings on human decision making, we propose SafeBehavior, a novel hierarchical jailbreak defense mechanism that simulates the adaptive multistage reasoning process of humans. SafeBehavior decomposes safety evaluation into three stages: intention inference to detect obvious input risks, self introspection to assess generated responses and assign confidence based judgments, and self revision to adaptively rewrite uncertain outputs while preserving user intent and enforcing safety constraints. We extensively evaluate SafeBehavior against five representative jailbreak attack types including optimization based, contextual manipulation, and prompt based attacks and compare it with seven state of the art defense baselines. Experimental results show that SafeBehavior significantly improves robustness and adaptability across diverse threat scenarios, offering an efficient and human inspired approach to safeguarding LLMs against jailbreak attempts. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_26345 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | SafeBehavior: Simulating Human-Like Multistage Reasoning to Mitigate Jailbreak Attacks in Large Language Models Zhao, Qinjian Wang, Jiaqi Gao, Zhiqiang Dou, Zhihao Abuhaija, Belal Huang, Kaizhu Artificial Intelligence Large Language Models (LLMs) have achieved impressive performance across diverse natural language processing tasks, but their growing power also amplifies potential risks such as jailbreak attacks that circumvent built-in safety mechanisms. Existing defenses including input paraphrasing, multi step evaluation, and safety expert models often suffer from high computational costs, limited generalization, or rigid workflows that fail to detect subtle malicious intent embedded in complex contexts. Inspired by cognitive science findings on human decision making, we propose SafeBehavior, a novel hierarchical jailbreak defense mechanism that simulates the adaptive multistage reasoning process of humans. SafeBehavior decomposes safety evaluation into three stages: intention inference to detect obvious input risks, self introspection to assess generated responses and assign confidence based judgments, and self revision to adaptively rewrite uncertain outputs while preserving user intent and enforcing safety constraints. We extensively evaluate SafeBehavior against five representative jailbreak attack types including optimization based, contextual manipulation, and prompt based attacks and compare it with seven state of the art defense baselines. Experimental results show that SafeBehavior significantly improves robustness and adaptability across diverse threat scenarios, offering an efficient and human inspired approach to safeguarding LLMs against jailbreak attempts. |
| title | SafeBehavior: Simulating Human-Like Multistage Reasoning to Mitigate Jailbreak Attacks in Large Language Models |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2509.26345 |