Witness Evidence Portfolios: Single-Prefill Risk Detection for Closed Multimodal Answers
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 produces the answer. Witness Evidence Portfolios (WEP) first estimates, layer by layer, which visual contributions support or contradict the predicted candidate. It summarizes these contributions through two interpretable route families: question-...
| Authors | Fexiang Liu, Shiye Wang, Qiang Qiu et al. |
|---|---|
| Primary category | cs.CV |
| Published | 30 July 2026 |
| Last updated at source | 30 July 2026 |
Silytics shows public metadata from arXiv and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.