Multimodal Agent MemorySubmit a paper

MemoryExplorer / LMEE-Bench

Explore with Long-term Memory: A Benchmark and Multimodal LLM-based Reinforcement Learning Framework for Embodied Exploration

Sen Wang, Bangwei Liu, Zhenkun Gao, Lizhuang Ma, Xuhong Wang, Yuan Xie, et al. · 2026-03-22

Read the paperPDF

Management
T2 Dynamic Memory
Memory changes during deployment through a stable management policy.
Primary modality
Embodied & 3D
Also: text
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. episodic / exploration memory

    1. Represented asSpatial, temporal & cross-modal structured memoryCross-represented explicit
    2. Stored inVector-indexed collection
    3. Retrieved bySimilarity search
    4. Source evidenceSource not retained
    Agent collects episodic memories during multi-goal navigation. episodic retrieval supplies memories to cognition/decision modules.Full-paper method / memory representation and retrieval · PDF pp. 2, 16–17 · Figure 1 / method overview

  2. episodic / exploration memory

    1. Represented asExperience-conditioned parametric memoryInternalized latent & parametric
    2. Stored inModular adapter bank, Vector-indexed collection
    3. Retrieved bySimilarity search, Latent / parametric activation
    4. Source evidenceSource not retained
    Agent collects episodic memories during multi-goal navigation. episodic retrieval supplies memories to cognition/decision modules.Full-paper method / memory representation and retrieval · PDF pp. 2, 16–17 · Figure 1 / method overview

How memory changes (T2 operations)

  • Update · Rule-based update

    During operation, MemoryExplorer / LMEE-Bench applies fixed-rule insertion or replacement to episodic / exploration memory; the memory state changes while the governing update policy remains fixed.

Related systems