Multimodal Agent MemorySubmit a paper

VisualMem

Personal Visual Memory from Explicit and Implicit Evidence

Viet Nguyen, Thao Nguyen, Vishal M. Patel, Yuheng Li · 2026-05-27

Read the paperPDF

Management
T2 Dynamic Memory
Memory changes during deployment through a stable management policy.
Primary modality
Image
Also: text
Representation
Hybrid
Distinct operative representation paths are explicitly linked for downstream retrieval, reasoning, or control.

How experience becomes memory

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

  1. personal image evidence

    1. Represented asRetained perceptual source evidenceSource- and modality-specific
    2. Stored inVector-indexed collection
    3. Retrieved bySimilarity search
    4. Source evidenceSource directly available
    Personal images are preserved rather than collapsed into generic captions.Full-paper method / memory representation and retrieval · PDF p. 3 · Figure 1

  2. structured personal visual memory

    1. Represented asSpatial, temporal & cross-modal structured memoryCross-represented explicit
    2. Stored inGraph / relational store
    3. Retrieved byStructure traversal, Similarity search
    4. Source evidenceSource linked & recoverable
    Conversational context resolves identity, ownership, and durable user facts into a structured visual memory.Full-paper method / memory representation and retrieval · PDF p. 3 · Figure 1

How memory changes (T2 operations)

  • Update · Prompted LLM/VLM record update

    During operation, VisualMem applies LLM-mediated memory editing to personal image evidence; the memory state changes while the governing update policy remains fixed.

  • Revision · Evidence-weighted belief revision

    Cross-turn evidence is reconciled to resolve identity, ownership, and persistent user facts. For VisualMem, this specifically supports evidence-weighted revision of stored belief or confidence of personal image evidence.

  • Revision · LLM-based conflict resolution

    During operation, VisualMem applies LLM-based resolution of conflicting records to personal image evidence; the memory state changes while the governing update policy remains fixed.

  • Consolidation · Merging & deduplication

    During operation, VisualMem applies merging or deduplication of overlapping entries to personal image evidence; the memory state changes while the governing update policy remains fixed.

Related systems