MiniVer-V: Identifying Minimal Sufficient Evidence for Short Video Verification
作者: Leran Chen, Lingnan Kong, Zile Cai
分类: cs.CV, cs.CL, cs.MM
发布日期: 2026-10-08
备注: 33 pages, 2 figures, 20 tables
💡 一句话要点
提出MiniVer-V以解决短视频验证中的证据充分性问题
🎯 匹配领域: 支柱九:具身大模型 (Embodied Foundation Models)
关键词: 短视频验证 证据充分性 多模态证据 贪婪搜索 事实核查 声明一致性 外部证据支持
📋 核心要点
- 现有短视频验证方法要么提供过多证据导致噪声,要么仅依赖主题相关性,未能有效识别证据的充分性。
- 本文提出MiniVer-V基准和两层验证框架,明确证据充分性作为选择标准,并通过贪婪搜索组装证据。
- 实验结果显示,使用MiniVer-V方法的Macro-F1达0.510,使用的证据单元仅为全证据集的16%,在不足案例识别上有显著提升。
📝 摘要(中文)
短视频事实核查中的核心挑战是识别哪些证据足以支持验证结论。现有方法要么提供所有可用证据,导致噪声干扰,要么通过主题相关性选择证据,混淆了相关性与充分性。本文提出MiniVer-V基准,包含195个短视频及三种判决注释(支持、反驳、不足),以及5,510个多模态证据单元。我们提出的两层验证框架将声明-视频一致性与事实判决分开,并通过贪婪搜索组装证据,直到达到充分性阈值。实验结果表明,该方法在效率和不足案例识别上显著优于传统方法。
🔬 方法详解
问题定义:本文旨在解决短视频验证中证据充分性识别的问题。现有方法往往提供过多的证据,导致信息噪声,或仅依赖于主题相关性,未能有效区分相关性与充分性。
核心思路:我们将证据的充分性作为选择标准,提出MiniVer-V基准,并设计了一个两层验证框架,分别评估声明-视频一致性和事实判决的外部证据支持。
技术框架:整体架构包括两个主要模块:第一层评估内部证据的声明-视频一致性,第二层进行事实判决,需外部证据的支持。通过贪婪搜索组装证据,直到达到充分性阈值。
关键创新:最重要的创新在于明确证据充分性作为选择标准,并通过贪婪搜索策略避免强制判决,允许在证据不足时输出“不足”结果。
关键设计:在实验中,外部证据被证明对事实判决至关重要,而内部视频证据则确保了声明与视频的一致性。实验还表明,使用的平均证据单元数为4.5,显著低于全证据集的27.7,但性能接近,显示出高效的证据利用能力。
🖼️ 关键图片
📊 实验亮点
实验结果显示,MiniVer-V方法在使用4.5个证据单元的情况下,Macro-F1达0.510,与全证据基线(0.518)在统计上无显著差异,同时在不足案例的识别上显著提升,验证了该方法的有效性和高效性。
🎯 应用场景
该研究在短视频事实核查、社交媒体内容验证等领域具有广泛的应用潜力。通过提高证据利用效率,能够帮助验证者更快速、准确地做出判断,减少误判和信息过载的风险,未来可能推动更智能的内容审核系统的发展。
📄 摘要(原文)
A core challenge in short-video fact-checking is identifying which evidence is sufficient to support a verification conclusion. Existing approaches either give the verifier all available evidence, introducing noise, or select evidence by topical relevance, which conflates relatedness with sufficiency. We identify evidential sufficiency as the selection criterion: whether a subset of evidence is adequate to support a confident verdict without redundancy. We introduce MiniVer-V, a benchmark of 195 short videos with three-way verdict annotations (supported, refuted, insufficient) and 5,510 multimodal evidence units spanning visual keyframes, speech transcripts, and web-retrieved external sources. We propose a two-layer verification framework that separates claim-video consistency, assessed from internal evidence, from factual verdict determination, which additionally requires external corroboration. On top of it, a sufficiency-driven greedy search assembles evidence until a sufficiency threshold is met and outputs insufficient when the candidate pool is exhausted, rather than forcing a verdict. With Claude Sonnet 4, the method reaches a Macro-F1 of 0.510 using 4.5 evidence units on average (16% of the full evidence set), statistically indistinguishable from the full-evidence baseline (0.518 with 27.7 units), while significantly improving recognition of insufficient cases over the same search without abstention. The efficiency result replicates with GPT-5.5 and holds only partially with an open-weight Qwen2.5-72B verifier. Ablations show that external evidence is indispensable for factual determination, while internal video evidence grounds the verdict in claim-video consistency. These findings suggest that evidence-efficient verification is achievable, and that explicit abstention is needed when evidence is genuinely inadequate.