研究人员推出了一种名为EarlyTom的新框架，旨在通过在视觉编码器早期压缩视觉令牌来提高视频大语言模型（Video-LLMs）的效率。该方法在不牺牲准确性的前提下，显著降低了首个令牌生成时间（TTFT）和计算成本。同时，OmniPro和VideoOdyssey等新基准正在开发中，用于评估全模态模型在理解流式和长上下文视频数据方面的先进能力，以解决现有评估方法的局限性。 AI
影响 这些进展旨在使Video-LLMs在实际应用中更加实用和高效，并为评估其复杂能力树立新标准。
排序理由 多篇研究论文介绍了用于视频理解的新框架和基准。
在 Hugging Face Daily Papers 阅读 →
AI 生成摘要 · Google Gemini · 来自 8 个来源。 我们如何撰写摘要 →
EarlyTom is a training-free framework that compresses visual tokens early in the vision encoder to reduce time-to-first-token and computational costs while maintaining model accuracy.
OmniPro is introduced as the first benchmark for evaluating omni-modal large language models' proactive streaming video understanding, featuring diverse tasks and dual-mode evaluation protocols.
arXiv:2605.30010v1 Announce Type: new Abstract: Video large language models (Video-LLMs) have demonstrated strong capabilities in video understanding tasks. However, their practical deployment is still hindered by the inefficiency introduced by processing massive amounts of visua…
Video large language models (Video-LLMs) have demonstrated strong capabilities in video understanding tasks. However, their practical deployment is still hindered by the inefficiency introduced by processing massive amounts of visual tokens. Although recent approaches achieve ext…
arXiv:2605.26584v1 Announce Type: new Abstract: Omnimodal large language models enable unified audio video understanding, but long joint token sequences make inference costly, and existing benchmarks do not fully isolate audio visual association in noisy user generated videos. We…
arXiv:2605.25621v1 Announce Type: new Abstract: While streaming omni-video understanding demands continuous perception and proactive, real-time interaction, this crucial area remains largely under-explored. Current omni-modal methods are inherently designed for offline settings, …
While streaming omni-video understanding demands continuous perception and proactive, real-time interaction, this crucial area remains largely under-explored. Current omni-modal methods are inherently designed for offline settings, limiting their applicability in streaming scenar…
arXiv:2605.22907v1 Announce Type: new Abstract: Real-world long video understanding requires models to perform continuous tracking, information integration and memory retention over massive temporal spans within extreme video durations. Mastering this intense cognitive load const…