SOURCE-LINKED INTELLIGENCE

Dynamic Alignment Compensation for Hallucination Mitigation in Large Vision-Language Models

arXiv · AI, language, vision and robotics · article · Aug 28, 2026 · UTC

Large Vision-Language Models (LVLMs) remain prone to hallucinations, producing responses that are irrelevant or inconsistent with the multimodal input. Existing mitigation methods mainly rely on external supervision, output calibration, or attention regulation, leaving the internal representation dynamics of autoregressive generation underexplored. We identify an inference-time failure mode in which cross-modal representations degrade across decoder layers and drift across generation steps, destabilizing token prediction and increasing hallucination risk. We propose \emph{Dynamic Alignment Com

Read original source ↗ Open in workspace

recordType

paper

region

Global