Multimodal Agent MemorySubmit a paper

StoryMem

StoryMem: Multi-shot Long Video Storytelling with Memory

Kaiwen Zhang, Liming Jiang, Angtian Wang, Jacob Zhiyuan Fang, Tiancheng Zhi, Qing Yan, et al. · 2025-12-22

Read the paperPDF

Management
T2 Dynamic Memory
Memory changes during deployment through a stable management policy.
Primary modality
Video & streaming
Also: text
Representation
Source- and modality-specific

How experience becomes memory

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

  1. historical keyframe memory bank

    1. Represented asRetained perceptual source evidenceSource- and modality-specific
    2. Stored inFlat sequential records
    3. Retrieved byMetadata / spatiotemporal retrieval
    4. Source evidenceSource directly available
    A compact bank stores selected keyframes from previously generated shots and injects them into later generation.Full-paper method / memory representation and retrieval · PDF p. 3 · Figure 1

How memory changes (T2 operations)

  • Update · Rule-based update

    During operation, StoryMem applies fixed-rule insertion or replacement to historical keyframe memory bank; the memory state changes while the governing update policy remains fixed.

  • Forgetting · Discrete eviction

    Only informative and aesthetically stable keyframes survive the compact-memory selection process. For StoryMem, this specifically supports explicit deletion or eviction of stored items of historical keyframe memory bank.

Related systems