نسخة أولية وصول مفتوح
Representation--Behavior Alignment for Explainable Weakly-Supervised Video Anomaly Detection
Multimodal Large Language Models (MLLMs) provide a natural way to make video anomaly detection more explainable. However, their final decisions do not always fully use the discriminative information contained in their hidden states, an issue we refer to as representation--behavior misalignment. We decompose this gap in …