构建交通视频问答基准，检验模型在相似场景下区分真危险与假假设的能力。
CCTVBench: Contrastive Consistency Traffic VideoQA Benchmark for Multimodal LLMs
- 用真实事故视频与生成的反事实视频配对，设计对比一致性的问答任务。
- 发现现有模型在识别假命题上表现差，常误判非危险为危险。
- 提出新推理方法，提升模型在真假判断上的可靠性，适合自动驾驶安全评估。
交通安全决策需要对比一致性：模型必须在事故发生时检测出真实危险，并在近乎相同的反事实场景中可靠排除看似合理但错误的假设。我们提出CCTVBench，一个基于真实事故视频与世界模型生成的反事实视频配对的交通视频问答基准，搭配最小差异、互斥的假设问题。该基准强制每组视频-问题四元组遵循单一结构化决策模式，并提供可操作的诊断，将失败分解为正向遗漏、正向混淆、负向幻觉和互斥性违反，同时区分视频与问题层面的一致性。在开源与专有视频大模型上的实验表明，标准实例级问答指标与四元组级对比一致性之间存在巨大且持续的差距，其中‘无一正确’选项的不可靠拒绝是关键瓶颈。最后，我们引入C-TCD，一种对比解码方法，在推理时使用语义互斥的反事实视频作为对比输入，同时提升实例级问答性能与对比一致性。
原文摘要 · Abstract (English)
Safety-critical traffic reasoning requires contrastive consistency: models must detect true hazards when an accident occurs, and reliably reject plausible-but-false hypotheses under near-identical counterfactual scenes. We present CCTVBench, a Contrastive Consistency Traffic VideoQA Benchmark built on paired real accident videos and world-model-generated counterfactual counterparts, together with minimally different, mutually exclusive hypothesis questions. CCTVBench enforces a single structured decision pattern over each video question quadruple and provides actionable diagnostics that decompose failures into positive omission, positive swap, negative hallucination, and mutual-exclusivity violation, while separating video versus question consistency. Experiments across open-source and proprietary video LLMs reveal a large and persistent gap between standard per-instance QA metrics and quadruple-level contrastive consistency, with unreliable none-of-the-above rejection as a key bottleneck. Finally, we introduce C-TCD, a contrastive decoding approach leveraging a semantically exclusive counterpart video as the contrast input at inference time, improving both instance-level QA and contrastive consistency.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。