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

TL;DR

M3-AD introduces a reflection-aware multimodal framework for industrial anomaly detection, enhancing reliability and robustness through self-correction mechanisms and comprehensive evaluation resources.

Contribution

It presents M3-AD, a unified framework with reflection-aware learning and new benchmark resources for systematic evaluation in industrial anomaly detection.

Findings

RA-Monitor outperforms existing MLLMs in zero-shot anomaly detection

Reflection-aware self-correction improves decision robustness

Extensive experiments validate the effectiveness of M3-AD

Abstract

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.