Despite the growing video understanding capabilities of recent Multimodal Large Language Models (MLLMs), existing video benchmarks primarily assess understanding based on models' static, internal knowledge, rather than their ability to learn and adapt from dynamic, novel contexts from few examples. To bridge this gap, we present Demo-driven Video In-Context Learning, a novel task focused on learning from in-context demonstrations to answer questions about the target videos. Alongside this, we propose Demo-ICL-Bench, a challenging benchmark designed to evaluate demo-driven video in-context learning capabilities. Demo-ICL-Bench is constructed from 1200 instructional YouTube videos with associated questions, from which two types of demonstrations are derived: (i) summarizing video subtitles for text demonstration; and (ii) corresponding instructional videos as video demonstrations. To effectively tackle this new challenge, we develop Demo-ICL, an MLLM with a two-stage training strategy: video-supervised fine-tuning and information-assisted direct preference optimization, jointly enhancing the model's ability to learn from in-context examples. Extensive experiments with state-of-the-art MLLMs confirm the difficulty of Demo-ICL-Bench, demonstrate the effectiveness of Demo-ICL, and thereby unveil future research directions.

翻译：尽管近年来多模态大语言模型（MLLMs）的视频理解能力不断增强，但现有的视频基准测试主要评估模型基于其静态内部知识进行理解的能力，而非其从少量示例的动态新颖上下文中学习和适应的能力。为弥补这一差距，我们提出了演示驱动的视频上下文学习，这是一项专注于从上下文演示中学习以回答关于目标视频问题的新任务。与此同时，我们提出了Demo-ICL-Bench，这是一个旨在评估演示驱动视频上下文学习能力的具有挑战性的基准。Demo-ICL-Bench构建自1200个带有相关问题的YouTube教学视频，从中衍生出两种类型的演示：（i）总结视频字幕作为文本演示；以及（ii）对应的教学视频作为视频演示。为有效应对这一新挑战，我们开发了Demo-ICL，这是一个采用两阶段训练策略的MLLM：视频监督微调和信息辅助直接偏好优化，共同增强了模型从上下文示例中学习的能力。与最先进MLLMs的大量实验证实了Demo-ICL-Bench的难度，证明了Demo-ICL的有效性，从而揭示了未来的研究方向。