To See or To Please: Uncovering Visual Sycophancy and Split Beliefs in VLMs 事件

To See or To Please: Uncovering Visual Sycophancy and Split Beliefs in VLMs arXiv:2603.18373v3 Announce Type: replace Abstract: When VLMs answer correctly, do they genuinely rely on visual information? We introduce a Tri-Layer Diagnostic Framework with three per-sample metrics: Latent Anomaly Detection, Visual Necessity Score, and Competition Score, which disentangle perception, dependency, and alignment failures. Across 9 VLMs and 9,000 model-sample pairs under counterfactual blind, noise, and