Work

Does Chain-of-Thought Reasoning Really Reduce Harmfulness from Jailbreaking?

Year: 2025

Type: conference-paper

Abstract: Jailbreak attacks have been observed to largely fail against recent reasoning models enhanced by Chain-of-Thought (CoT) reasoning.However, the underlying mechanism remains underexplored, and relying solely on reasoning capacity may raise security concerns.In this paper, we try to answer the question: Does CoT reasoning really reduce harmfulness from jailbreaking?Through rigorous theoretical analysis, we demonstrate that CoT reasoning has dual effects on jailbreaking harmfulness.Based on the theoretical insights, we propose a novel jailbreak method, FICDETAIL, whose practical performance validates our theoretical findings.Warning: This work contains potentially offensive LLM-generated content.

DOI: 10.18653/v1/2025.findings-acl.339

Authors: Chengda Lu, Xiaoyu Fan, Yu Huang, Rongwu Xu, Jijie Li, Wei Xu

Institutions: Shanghai Qi Zhi Institute, Tsinghua University, University of Pennsylvania, Beijing Academy of Artificial Intelligence

Language: English

Topic: Adversarial Robustness in Machine Learning

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: gold

Funder: National Natural Science Foundation of China

Awards: G366789873, G7422017672

OpenAlex ID: W4412888656