研究人员开发了新的框架来改进AI模型理解和推理长视频的方式。Homer利用分层记忆结构，按时序和因果联系组织信息，实现更复杂的叙事推理。Latent-VC通过使用循环缓存来在生成过程中保留视觉记忆，解决了“视觉锚定衰减”问题，从而产生更准确、更简洁的响应。EGAgent专注于以自我为中心的视频理解，应用于可穿戴AI助手等场景，采用实体场景图和规划代理进行长时间的多跳推理。 AI
影响 这些进展可以实现更复杂的AI应用，用于分析扩展视频内容，例如在监控、内容摘要和个人AI助手领域。
排序理由 三篇不同的研究论文提出了长视频理解的新颖框架。
AI 生成摘要 · Google Gemini · 来自 4 个来源。 我们如何撰写摘要 →
arXiv:2607.02588v1 Announce Type: cross Abstract: Multimodal large language models excel on short clips but struggle on hour-long videos in an online setting, where frames are processed incrementally under limited memory. Existing online methods either retain compact visual repre…
arXiv:2607.02607v1 Announce Type: cross Abstract: Video reasoning requires Large Multimodal Models (LMMs) to remain grounded in dense evidence, yet existing systems largely adopt "read-once, generate-many" paradigm, in which visual grounding weakens during generation. This phenom…
arXiv:2601.18157v3 Announce Type: replace-cross Abstract: The advent of always-on personal AI assistants, enabled by all-day wearable devices such as smart glasses, demands a new level of contextual understanding, one that goes beyond short, isolated events to encompass the conti…
arXiv:2607.04559v1 Announce Type: new Abstract: The performance of vision-language models (VLMs) in video understanding declines with increasing video duration, as video moments unrelated to the query confuse their language components. Multimodal retrieval has emerged as a critic…