M3-AD: Reflection-aware Multi-modal, Multi-category, and Multi-dimensional Benchmark and Framework for Industrial Anomaly Detection

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Huang, Chao, Li, Yanhui, Cao, Yunkang, Wang, Wei, Huang, Hongxi, Wen, Jie, Ren, Wenqi, Cao, Xiaochun
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908856804704256
author Huang, Chao
Li, Yanhui
Cao, Yunkang
Wang, Wei
Huang, Hongxi
Wen, Jie
Ren, Wenqi
Cao, Xiaochun
author_facet Huang, Chao
Li, Yanhui
Cao, Yunkang
Wang, Wei
Huang, Hongxi
Wen, Jie
Ren, Wenqi
Cao, Xiaochun
contents Although multimodal large language models (MLLMs) have advanced industrial anomaly detection toward a zero-shot paradigm, they still tend to produce high-confidence yet unreliable decisions in fine-grained and structurally complex industrial scenarios, and lack effective self-corrective mechanisms. To address this issue, we propose M3-AD, a unified reflection-aware multimodal framework for industrial anomaly detection. M3-AD comprises two complementary data resources: M3-AD-FT, designed for reflection-aligned fine-tuning, and M3-AD-Bench, designed for systematic cross-category evaluation, together providing a foundation for reflection-aware learning and reliability assessment. Building upon this foundation, we propose RA-Monitor, which models reflection as a learnable decision revision process and guides models to perform controlled self-correction when initial judgments are unreliable, thereby improving decision robustness. Extensive experiments conducted on M3-AD-Bench demonstrate that RA-Monitor outperforms multiple open-source and commercial MLLMs in zero-shot anomaly detection and anomaly analysis tasks. Code will be released at https://github.com/Yanhui-Lee/M3-AD.
format Preprint
id arxiv_https___arxiv_org_abs_2603_00055
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle M3-AD: Reflection-aware Multi-modal, Multi-category, and Multi-dimensional Benchmark and Framework for Industrial Anomaly Detection
Huang, Chao
Li, Yanhui
Cao, Yunkang
Wang, Wei
Huang, Hongxi
Wen, Jie
Ren, Wenqi
Cao, Xiaochun
Machine Learning
Artificial Intelligence
Although multimodal large language models (MLLMs) have advanced industrial anomaly detection toward a zero-shot paradigm, they still tend to produce high-confidence yet unreliable decisions in fine-grained and structurally complex industrial scenarios, and lack effective self-corrective mechanisms. To address this issue, we propose M3-AD, a unified reflection-aware multimodal framework for industrial anomaly detection. M3-AD comprises two complementary data resources: M3-AD-FT, designed for reflection-aligned fine-tuning, and M3-AD-Bench, designed for systematic cross-category evaluation, together providing a foundation for reflection-aware learning and reliability assessment. Building upon this foundation, we propose RA-Monitor, which models reflection as a learnable decision revision process and guides models to perform controlled self-correction when initial judgments are unreliable, thereby improving decision robustness. Extensive experiments conducted on M3-AD-Bench demonstrate that RA-Monitor outperforms multiple open-source and commercial MLLMs in zero-shot anomaly detection and anomaly analysis tasks. Code will be released at https://github.com/Yanhui-Lee/M3-AD.
title M3-AD: Reflection-aware Multi-modal, Multi-category, and Multi-dimensional Benchmark and Framework for Industrial Anomaly Detection
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2603.00055