Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

ReShift: Aha-Moment-Driven Reasoning-Level Backdoor Attacks on Vision-Language Models

About

Vision--Language Models (VLMs) are increasingly deployed in safety-critical applications, yet remain vulnerable to backdoor attacks. Existing methods primarily manipulate final outputs, often producing reasoning traces that are inconsistent or easily detectable. In this paper, we propose ReShift, the novel aha-moment-driven reasoning-level backdoor framework that explicitly redirects the internal chain-of-thought (CoT) trajectory while preserving surface-level coherence. ReShift introduces a Poisoned Reasoning-Aware Data Construction (PRDC) pipeline and a Supervised--Reinforcement Joint Optimization (SRJO) strategy to induce stable trigger-conditioned reasoning shifts. We further formalize Entropy Rebound as a principled signal for characterizing reasoning redirection and provide theoretical guaranties linking entropy gaps to trajectory-level divergence. Extensive experiments demonstrate that ReShift achieves high attack success rates while maintaining clean-task performance and realistic reasoning traces, substantially improving stealthiness against existing defenses.

Zhihao Dou, Qinjian Zhao, Zhiqiang Gao, Sumon Biswas• 2026

Related benchmarks

TaskDatasetResultRank
Visual Question AnsweringA-OKVQA
Acc90
240
Mathematical ReasoningMathVista
Accuracy49
22
Visual Question AnsweringScienceQA
Coherence4.1
10
Mathematical ReasoningMathVista
Coherence3.02
10
Multi-discipline Multi-modal UnderstandingMMMU
Coherence3.67
10
Backdoor DetectionA-OKVQA
Detection Accuracy (DACC)13
8
Backdoor DetectionScienceQA
DACC17
8
Backdoor DetectionMMMU
Detection Accuracy17
8
Backdoor DetectionMathVista
DACC16
8
Mathematical ReasoningMathVision
Coherence3.22
5
Showing 10 of 10 rows

Other info

Follow for update