学习拒绝：多模态大语言模型中的动作否定
Learning to Deny: Action Denial in Multimodal Large Language Models
AI总结：
提出UCF101-AD基准测试，评估多模态大语言模型在强上下文线索下识别动作缺失的能力，发现模型倾向于肯定动作而非验证其真实性，并通过因果图CausalAct减少误报。
摘要：
多模态大语言模型（MLLMs）在视频理解方面取得了快速进展，在标准基准测试中实现了强大的零样本和少样本识别。然而，它们在面对强烈上下文线索时，识别活动未发生（即否定动作）的能力仍未被充分探索。我们引入了UCF101-AD，一个大规模基准测试，由成对的“动作存在”和“动作否定”片段组成，旨在评估这种否定能力。UCF101-AD中的每个负视频保留了与其正视频相同的上下文和运动线索，包括人物、物体和位置，但定义性的动作本身明确缺失。评估20个最先进的MLLMs揭示了一个一致的失败：在正动作类上准确率超过85%的模型，在其动作否定对应类上准确率骤降至50%以下，表明模型倾向于确认看似合理的动作，而非验证其是否真实发生。这暴露了现代视频理解中的一个关键盲点：无法因果推理一个动作是否实际发生。为了探究这个问题，我们探索了一个因果图公式CausalAct，它通过自然语言提示表达场景结构，将上下文、交互和运动联系起来。融入这种因果线索显著减少了误报，表明否定是一种可学习的推理技能。UCF101-AD为诊断和改进多模态模型中的因果推理提供了新视角。数据集和相关代码：此 https URL。
英文摘要：
Multimodal large language models (MLLMs) have rapidly advanced video understanding, achieving strong zero-shot and few-shot recognition across standard benchmarks. Yet their ability to deny an action by recognizing when an activity is not happening despite strong contextual cues remains largely unexplored. We introduce UCF101-AD, a large-scale benchmark consisting of paired Action-Presence and Action-Denial clips, designed to evaluate this capacity for denial. Each negative video in UCF101-AD preserves the same contextual and motion cues, including persons, objects, and locations, as its positive counterpart, but the defining action itself is explicitly absent. Evaluating 20 state-of-the-art MLLMs reveals a consistent failure: models that exceed 85% accuracy on the positive action classes collapse below 50% on their action-denial counterparts, indicating a strong inclination to affirm plausible actions rather than verify that they truly occur. This exposes a critical blind spot in modern video understanding: the inability to reason causally about whether a motion actually happens. To probe this issue, we explore a causal graph formulation, CausalAct, which expresses scene structure through natural-language prompts linking context, interaction, and motion. Incorporating such causal cues substantially reduces false positives, demonstrating that denial is a learnable reasoning skill. UCF101-AD provides a new lens for diagnosing and improving causal reasoning in multimodal models. Dataset and relevant code: https://github.com/raiyaan-abdullah/Learn-to-Deny.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
