首个评估多流视频理解的基准，揭示大模型并发处理能力短板。
X-Stream: Exploring MLLMs as Multiplexers for Multi-Stream Understanding
- 将多模态大模型视为信号复用器，测试其跨流推理能力。
- 实测顶尖模型在多流场景下仅得50%分数，主动响应能力差。
- 适合研究多流交互、视频理解或实时系统设计的学者。
尽管视频流理解已取得显著进展，但实际应用如直播体育转播、自动驾驶和多屏协作，本质上需要持续的多流交互。然而，现有基准仍局限于单流范式，难以评估在线跨流推理能力。为此，我们提出X-Stream，首个专注于多流流理解的基准。该数据集包含4,220个精心标注的问答对，覆盖932个视频，涵盖11个子任务，涉及多窗口、多视角和多设备场景。关键的是，数据集通过新型双验证流程构建，防止对单一流的过度依赖。我们首次将多模态大语言模型（MLLMs）概念化为原始复用器，并基于信号复用理论系统评估其性能。大规模在线推理实验揭示：当前最先进的MLLMs在并发流中表现不佳，平均得分仅约50%，且缺乏主动应对能力。最终，X-Stream揭示了现有复用方案的权衡，为下一代多流智能体提供实践评估框架与实证指导。
原文摘要 · Abstract (English)
While video streaming understanding has made significant strides, real-world applications, such as live sports broadcasting, autonomous driving, and multi-screen collaboration, inherently demand continuous, multi-stream interactions. However, existing benchmarks are confined to single-stream paradigms, leaving a critical gap in evaluating online, cross-stream reasoning. To bridge this, we introduce X-Stream, the first benchmark dedicated to multi-stream streaming understanding. Comprising 4,220 rigorously curated QA pairs across 932 videos, X-Stream evaluates 11 subtasks across multi-window, multi-view, and multi-device scenarios. Crucially, our dataset is constructed using a novel dual-verification pipeline that prevents over-reliance on a single stream. Furthermore, we pioneer the conceptualization of multi-modal large language models (MLLMs) as naive multiplexers, systematically evaluating their performance through the lens of Signal Multiplexing Theory. Our extensive online inference experiments reveal a stark reality: state-of-the-art MLLMs struggle significantly with concurrent streams, achieving only about 50% score and exhibiting poor proactive ability. Ultimately, X-Stream exposes the trade-off of current multiplexing schemes, providing both a practical evaluation protocol and empirical guidance for next-generation multi-stream agents.
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。