Multimodal Agent MemorySubmit a paper

MuSEAgent

MuSEAgent: Multimodal Semantic Experience Agent

Shijian Wang, Jiarui Jin, Runhao Fu, Zexuan Yan, Xingjian Wang, Mengkang Hu, et al. · 2026-03-29

Read the paperPDF

Management
T1 Static Memory
Memory is built once and reused without change during deployment.
Primary modality
Image
Also: text
Representation
Cross-represented explicit

How experience becomes memory

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

  1. quality-filtered multimodal experience bank

    1. Represented asSpatial, temporal & cross-modal structured memoryCross-represented explicit
    2. Stored inVector-indexed collection
    3. Retrieved bySimilarity search, Planner / tool-assisted retrieval
    4. Source evidenceSource linked & recoverable
    Atomic decisions are abstracted into reusable multimodal experiences and retrieved through wide/deep semantic search.Full-paper method / memory representation and retrieval · PDF p. 4 · Figure 1

Related systems