Multimodal Agent MemorySubmit a paper

S3Mem

S3Mem: Structured Spatiotemporal Scene-Event Memory for Long-Horizon Interactive Question Answering

Encheng Su, Jinouwen Zhang, Jianyu Wu, Qiucheng Yu, Chen Tang, Pengze Li, et al. · 2026-05-28

Read the paperPDF

Management
T1 Static Memory
Memory is built once and reused without change during deployment.
Primary modality
Embodied & 3D
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. structured scene-event episodic memory

    1. Represented asSpatial, temporal & cross-modal structured memoryCross-represented explicit
    2. Stored inGraph / relational store
    3. Retrieved byMetadata / spatiotemporal retrieval, Structure traversal
    4. Source evidenceSource linked & recoverable
    Trajectories are written into structured scene-event units and routed through anchor-sensitive retrieval.Full paper: Structured Memory Writer · PDF p. 4 · Figure 1

Related systems