Witness Evidence Portfolios: Single-Prefill Risk Detection for Closed Multimodal Answers

arXiv:2607.27667v1 Announce Type: new Abstract: Reliable deployment of multimodal large language models (MLLMs) requires deciding whether a confident visual answer should be trusted, reviewed, or routed to a stronger system. Confidence scores capture candidate margins, but not where the estimated signed visual readouts associated with those margins come from or how they are distributed. We study inference-time risk detection for closed visual answers using the same white-box prefill path that pr...