arXiv · 2611.22673
FVScore: Specification-Driven Neuro-Symbolic Video Evaluation
Abstract
We introduce Faithfulness Video Score (FVSCORE), a neuro-symbolic framework for specification-driven video generation evaluation. Existing evaluators rely on holistic MLLM judgments or vision-language metrics, yet aligning their scores with human judgments and tracing them to specific temporal violations remain open challenges. FVSCORE addresses these limitations by compiling each video into explicit propositions capturing prompt requirements and implicit propositions capturing unstated physical, temporal, and cross-modal constraints. It grounds each proposition using specialized perceptual tools and evaluates the resulting confidence traces with soft temporal operators inspired by Linear Temporal Logic. A Significance-Aware Calibration step then aligns verification with human perception by retaining only perceptually significant violations before producing a timestamp- and evidence-grounded report. We conduct extensive experiments across public and complex multi-event benchmarks with the strongest video models to date, demonstrating that FVSCORE outperforms holistic and agent-based evaluators in human correlation while providing highly accurate and interpretable feedback.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Do Xuan Long, Long T. Le, Yiwen Song, Min-Yen Kan, Tomas Pfister, Yale Song. 2026-11-17. FVScore: Specification-Driven Neuro-Symbolic Video Evaluation. https://searcharxiv.com/abs/2611.22673
Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite Searcharxiv by Indiciae: BibTeX · RIS.