Multimodal Agent MemorySubmit a paper

FilmAgent

FilmAgent: A Multi-Agent Framework for End-to-End Film Automation in Virtual 3D Spaces

Zhenran Xu, Longyue Wang, Jifang Wang, Zhouyi Li, Senbao Shi, Xue Yang, et al. · 2025-01-22

Read the paperPDF

Management
T2 Dynamic Memory
Memory changes during deployment through a stable management policy.
Primary modality
Video & streaming
Also: text, Embodied & 3D
Representation
Hybrid
Distinct operative representation paths are deliberately linked or jointly consumed by downstream reasoning, generation, or control.

How experience becomes memory

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

  1. Persistent 3D world state (sets, characters, positions)

    1. Represented asModality-specific encodings & geometric stateSource- and modality-specific
    2. Stored inFlat sequential records
    3. Retrieved byPlanner / tool-assisted retrieval
    4. Source evidenceSource not retained
    Unity 3D scene state. Shots are staged in the persistent space; continuity is automatic.Full-paper method / memory representation and retrieval · PDF p. 2 · Figure 1 / method overview

  2. Persistent 3D world state (sets, characters, positions)

    1. Represented asSpatial, temporal & cross-modal structured memoryCross-represented explicit
    2. Stored inFlat sequential records
    3. Retrieved byPlanner / tool-assisted retrieval
    4. Source evidenceSource not retained
    Unity 3D scene state. Shots are staged in the persistent space; continuity is automatic.Full-paper method / memory representation and retrieval · PDF p. 2 · Figure 1 / method overview

How memory changes (T2 operations)

  • Update · Prompted LLM/VLM record update

    During operation, FilmAgent applies LLM-mediated memory editing to Persistent 3D world state (sets, characters, positions); the memory state changes while the governing update policy remains fixed.

  • Revision · Rule-based state correction

    During operation, FilmAgent applies fixed-rule correction of stored state to Persistent 3D world state (sets, characters, positions); the memory state changes while the governing update policy remains fixed.

Related systems