Multimodal Agent MemorySubmit a paper

AnomalyAgent

AnomalyAgent: Agentic Multimodal Reasoning for Few-Shot Anomaly Detection

Yi Zhang, Jiawen Zhu, Lele Fu, Guansong Pang · 2026-05-28

Read the paperPDF

Management
T2 Dynamic Memory
Memory changes during deployment through a stable management policy.
Primary modality
Image
Also: text
Representation
Source- and modality-specific

How experience becomes memory

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

  1. few-shot visual reference memory

    1. Represented asRetained perceptual source evidenceSource- and modality-specific
    2. Stored inVector-indexed collection
    3. Retrieved bySimilarity search
    4. Source evidenceSource directly available
    A fixed bank of few-shot visual references is retrieved to ground anomaly reasoning.Full paper: Memory-Augmented Self Calibration with Normal Reference Examples · PDF pp. 5–6 · Figure 1

How memory changes (T2 operations)

  • Revision · Evidence-weighted belief revision

    During operation, AnomalyAgent applies evidence-weighted revision of stored belief or confidence to few-shot visual reference memory; the memory state changes while the governing update policy remains fixed.

  • Update · Prompted LLM/VLM record update

    During operation, AnomalyAgent applies LLM-mediated memory editing to few-shot visual reference memory; the memory state changes while the governing update policy remains fixed.

Related systems