Multimodal Agent MemorySubmit a paper

MemoryVLA++

MemoryVLA++: Temporal Modeling via Memory and Imagination in Vision-Language-Action Models

Hao Shi, Weiye Li, Bin Xie, Yulin Wang, Renping Zhou, Tiancai Wang, et al. · 2026-06-08

Read the paperPDF

Management
T2 Dynamic Memory
Memory changes during deployment through a stable management policy.
Primary modality
Embodied & 3D
Also: Image, text
Representation
Internalized latent & parametric

How experience becomes memory

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

  1. perceptual-cognitive history and imagined future latents

    1. Represented asPersistent multimodal latent stateInternalized latent & parametric
    2. Stored inVector-indexed collection
    3. Retrieved bySimilarity search, Latent / parametric activation
    4. Source evidenceSource not retained
    The bank retrieves historical tokens while a world model supplies future latents.Full paper: Cognitive Memory · PDF p. 5 · Figure 1

How memory changes (T2 operations)

  • Update · Rule-based update

    During operation, MemoryVLA++ applies fixed-rule insertion or replacement to perceptual-cognitive history and imagined future latents; the memory state changes while the governing update policy remains fixed.

  • Consolidation · Merging & deduplication

    Redundancy-aware consolidation updates the bank. For MemoryVLA++, this specifically supports merging or deduplication of overlapping entries of perceptual-cognitive history and imagined future latents.

Related systems