Saved in:
Bibliographic Details
Main Authors: Miao, Junwen, Du, Penghui, Fan, Yingying, Liu, Yi, Wang, Yu, He, Runze, Huang, Lida, Wang, Yan
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.13671
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908967185154048
author Miao, Junwen
Du, Penghui
Fan, Yingying
Liu, Yi
Wang, Yu
He, Runze
Huang, Lida
Wang, Yan
author_facet Miao, Junwen
Du, Penghui
Fan, Yingying
Liu, Yi
Wang, Yu
He, Runze
Huang, Lida
Wang, Yan
contents Industrial anomaly detection (IAD) is challenging due to the subtle and highly localized nature of many defects, which single-pass vision--language models (VLMs) often fail to capture. Moreover, existing approaches lack mechanisms to actively acquire complementary evidence during inference. We propose AgentIAD, an agentic vision--language framework that enables iterative industrial inspection through a unified action space. The agent dynamically accesses two forms of memory during inspection: visual memory via the Perceptive Zoomer (PZ) for fine-grained local analysis, and retrieved memory via the Web Searcher (WS) and Comparative Retriever (CR) for external knowledge acquisition and cross-instance verification. This design allows the model to progressively gather evidence through multi-round perception--action reasoning. To effectively learn such policies under sparse supervision, AgentIAD adopts a two-stage training strategy: tool-aware supervised fine-tuning first initializes structured reasoning and memory-access behaviors, followed by agentic reinforcement learning to refine long-horizon decision policies. Extensive experiments show that, under the same backbone, AgentIAD improves classification accuracy by 5.92% over the previous state-of-the-art method on the MMAD benchmark while providing more reliable and interpretable anomaly analysis.
format Preprint
id arxiv_https___arxiv_org_abs_2512_13671
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle AgentIAD: Agentic Industrial Anomaly Detection via Adaptive Memory Augmentation
Miao, Junwen
Du, Penghui
Fan, Yingying
Liu, Yi
Wang, Yu
He, Runze
Huang, Lida
Wang, Yan
Computer Vision and Pattern Recognition
Industrial anomaly detection (IAD) is challenging due to the subtle and highly localized nature of many defects, which single-pass vision--language models (VLMs) often fail to capture. Moreover, existing approaches lack mechanisms to actively acquire complementary evidence during inference. We propose AgentIAD, an agentic vision--language framework that enables iterative industrial inspection through a unified action space. The agent dynamically accesses two forms of memory during inspection: visual memory via the Perceptive Zoomer (PZ) for fine-grained local analysis, and retrieved memory via the Web Searcher (WS) and Comparative Retriever (CR) for external knowledge acquisition and cross-instance verification. This design allows the model to progressively gather evidence through multi-round perception--action reasoning. To effectively learn such policies under sparse supervision, AgentIAD adopts a two-stage training strategy: tool-aware supervised fine-tuning first initializes structured reasoning and memory-access behaviors, followed by agentic reinforcement learning to refine long-horizon decision policies. Extensive experiments show that, under the same backbone, AgentIAD improves classification accuracy by 5.92% over the previous state-of-the-art method on the MMAD benchmark while providing more reliable and interpretable anomaly analysis.
title AgentIAD: Agentic Industrial Anomaly Detection via Adaptive Memory Augmentation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.13671