跳到正文
arXiv cs.AI· Do Xuan Long, Long T. Le, Yiwen Song, Min-Yen Kan, Tomas Pfister, Yale Song·· 2 小时前AI 评分28

FVScore:规格驱动的神经符号视频生成评测框架

FVScore: Specification-Driven Neuro-Symbolic Video Evaluation

AI 导读

FVSCORE 是一个神经符号视频生成评测框架,将视频编译为显式与隐式命题,用专门的感知工具定位,并借助受线性时序逻辑启发的软时序算子评估置信度轨迹,再通过显著性感知校准保留感知上显著的违规,生成带时间戳和证据的报告。在公开及复杂多事件基准上的实验显示,FVSCORE 与人类评分的相关性优于整体式和基于智能体的评测器,且反馈更准确、可解释。

正文 · 原文

View PDF HTML (experimental)

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.
Comments: Project page: this https URL
Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
Cite as: arXiv:2611.22673 [cs.CV]
  (or arXiv:2611.22673v1 [cs.CV] for this version)
  https://doi.org/10.48550/arXiv.2611.22673

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Long Do Xuan [view email]
[v1] Tue, 17 Nov 2026 12:25:33 UTC (7,033 KB)

来源:arXiv cs.AI · arxiv.org