AIIC AI Intelligence Centre

SOURCE-LINKED INTELLIGENCE

Target-Checked Reliability Score Refinement for Video Question Answering

arXiv · AI, language, vision and robotics · article · Sep 9, 2026 · UTC

Video-language models can answer multiple-choice questions with high confidence yet be wrong. We study whether answer-level reliability scores can be improved under target shift without retraining the models or changing their answers. We collect option-probability lists from three fixed video-language models under four deterministic video samplings and represent cross-view changes and cross-model agreement as a response graph. Using a labeled target pilot, we compare the original score, defined as the probability assigned to the chosen answer, with a histogram-based gradient-boosting (HGB) sco

Read original source ↗ Open in workspace

recordType
paper
region
Global

Evidence & attribution

First collected: 2026-09-20T19:52:05.078Z. This is not the publication date.