SOURCE-LINKED INTELLIGENCE

Not All Tokens Are Equal: Region-Aware Consistency Repair of Backdoors in MLLMs

arXiv · AI, language, vision and robotics · article · Aug 25, 2026 · UTC

MLLMs are increasingly deployed in user-facing applications, yet they inherit backdoor risks from the pipelines used to construct them: triggers may reside in images, texts, or both. Existing model-level backdoor removal methods, largely designed for conventional classifiers, show limited effectiveness on MLLMs, while MLLM-specific defenses mainly operate at inference time, filtering suspicious inputs without removing the backdoor embedded in the model. To address this gap and eliminate latent backdoors from MLLMs at their source, we present RACER, a model-level repair framework motivated by a

Read original source ↗ Open in workspace

recordType

paper

region

Global