长视频理解一直受限于 token 爆炸和注意力稀释，MemDreamer 用智能体检索和分层记忆解决了这个痛点。做视频分析、多模态研究的团队可以直接参考其框架，在现有模型上即插即用，值得一试。
MemDreamer 是一种新型框架，通过解耦感知与推理，将长视频理解转化为智能体探索过程。它采用分层图记忆架构，将视频流增量构建为三层语义抽象，并利用智能体工具增强检索机制，在推理时通过观察-推理-行动循环导航节点和逻辑边。实验表明，MemDreamer 在四个主流基准测试中达到最先进水平，与人类专家的差距缩小至仅 3.7 分。它仅使用全上下文 2% 的推理窗口，却带来 12.5 分的绝对准确率提升。此外，统计发现视觉语言模型在逻辑推理与长视频理解性能间存在强正线性相关，表明智能体能力扩展是多模态理解的新范式。
MemDreamer: Decoupling Perception and Reasoning for Long Video Understanding via Hierarchical Graph Memory and Agentic Retrieval Mechanism
Current Vision-Language Models struggle with hours-long videos because processing full-length visual sequences induces prohibitive token explosion and attention dilution. To overcome this, we introduce MemDreamer to decouple perception and reasoning, shifting long-video understanding into an agentic exploration process. As a plug-and-play framework, it incrementally streams videos to construct a Hierarchical Graph Memory, a top-down three-tier architecture for semantic abstraction, anchored by a foundational graph capturing spatiotemporal and causal relations. During inference, the reasoning model employs agentic tool-augmented retrieval, navigating hierarchies, searching nodes, and traversing logical edges via an Observation-Reason-Action loop. Experiments show MemDreamer achieves SOTA results across four mainstream benchmarks, narrowing the gap with human experts to only 3.7 points. It constrains the reasoning context window to merely 2% of full-context ingestion while delivering a 12.5 point absolute accuracy gain. Furthermore, statistical analysis uncovers a strong positive linear correlation between an VLM's performance on logic reasoning and long-video understanding benchmarks, establishing agentic capability scaling as a new paradigm for multimodal comprehension.