Multimodal Agent MemorySubmit a paper

VideoARM

VideoARM: Adaptive Reasoning and Memory for Long Video Understanding

Yufei Yin, Qianke Meng, Minghao Chen, Jiajun Ding, Zhenwei Shao, Zhou Yu · 2025-12-13

Read the paperPDF

Management
T2 Dynamic Memory
Memory changes during deployment through a stable management policy.
Primary modality
Video & streaming
Also: text
Representation
Hybrid
Distinct operative representation paths are explicitly linked for downstream retrieval, reasoning, or control.

How experience becomes memory

Each path is one memory component, traced from what is stored to how it is found again.

  1. multilevel visual clue memory

    1. Represented asRetained perceptual source evidenceSource- and modality-specific
    2. Stored inHierarchical multi-tier store
    3. Retrieved byMetadata / spatiotemporal retrieval, Planner / tool-assisted retrieval
    4. Source evidenceSource directly available
    The observe-think-act-memorize loop retains and updates multilevel visual clues.Full-paper method / memory representation and retrieval · PDF p. 2 · Figure 1

  2. hierarchical semantic memory

    1. Represented asPerception-to-text records & summariesCross-represented explicit
    2. Stored inHierarchical multi-tier store
    3. Retrieved byPlanner / tool-assisted retrieval
    4. Source evidenceSource linked & recoverable
    Hierarchical summaries provide long-range evidence to later reasoning steps.Full paper: 3. Method · PDF p. 3 · Figure 1

How memory changes (T2 operations)

  • Consolidation · Episodic / hierarchical summarization

    A fixed controller consolidates observations into multilevel clues. For VideoARM, this specifically supports episodic or hierarchical summarization of multilevel visual clue memory.

  • Update · Prompted LLM/VLM record update

    The memorize step updates retained clues during video processing. For VideoARM, this specifically supports LLM-mediated memory editing of multilevel visual clue memory.

Related systems