M3-AD: Reflection-aware Multi-modal, Multi-category, and Multi-dimensional Benchmark and Framework for Industrial Anomaly Detection
Not on the map. Why?
Abstract (arXiv)
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.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 36 of its 56 references that are in the map. First posted on arXiv 10 Feb 2026.
Builds on
36 of 56 references.
References · 36
- 2024Qwen2 Technical Reportcited by 2,076
- 2022Visual Prompt Tuningcited by 1,674
- 2021Towards Total Recall in Industrial Anomaly Detectioncited by 772
- 2025Vision-R1: Incentivizing Reasoning Capability in Multimodal Large Language Modelscited by 653
- 2020PaDiM: a Patch Distribution Modeling Framework for Anomaly Detection and Localizationcited by 591
- 2025VLM-R1: A Stable and Generalizable R1-style Large Vision-Language Modelcited by 515
- 2024Training Language Models to Self-Correct via Reinforcement Learningcited by 405
- 2022SPot-the-Difference Self-Supervised Pre-training for Anomaly Detection and Segmentationcited by 397
- 2025VL-Rethinker: Incentivizing Self-Reflection of Vision-Language Models with Reinforcement Learningcited by 315
- 2023WinCLIP: Zero-/Few-Shot Anomaly Classification and Segmentationcited by 311
- 2023AnomalyCLIP: Object-agnostic Prompt Learning for Zero-shot Anomaly Detectioncited by 270
- 2021VT-ADL: A Vision Transformer Network for Image Anomaly Detection and Localizationcited by 222
24 more
- 2023AnomalyGPT: Detecting Industrial Anomalies Using Large Vision-Language Modelscited by 169
- 2025DAST: Difficulty-Adaptive Slow-Thinking for Large Reasoning Modelscited by 154
- 2023Deep Industrial Image Anomaly Detection: A Surveycited by 147
- 2019Segmentation-Based Deep-Learning Approach for Surface-Defect Detectioncited by 145
- 2024AdaCLIP: Adapting CLIP with Hybrid Learnable Prompts for Zero-Shot Anomaly Detectioncited by 126
- 2021Mixed supervision for surface-defect detection: from weakly to fully supervised learningcited by 101
- 2025Visual Planning: Let's Think Only with Imagescited by 85
- 2025Towards Zero-Shot Anomaly Detection and Reasoning with Multimodal Large Language Modelscited by 54
- 2025SRPO: Enhancing Multimodal LLM Reasoning via Reflection-Aware Reinforcement Learningcited by 53
- 2024FiLo: Zero-Shot Anomaly Detection by Fine-Grained Description and High-Quality Localizationcited by 51
- 2024MMAD: A Comprehensive Benchmark for Multimodal Large Language Models in Industrial Anomaly Detectioncited by 46
- 2023Myriad: Large Multimodal Model by Applying Vision Experts for Industrial Anomaly Detectioncited by 39
- 2025Reflect, Retry, Reward: Self-Improving LLMs via Reinforcement Learningcited by 32
- 2025Bayesian Prompt Flow Learning for Zero-Shot Anomaly Detectioncited by 31
- 2025AnomalyR1: A GRPO-based End-to-end MLLM for Industrial Anomaly Detectioncited by 28
- 2025AdaptCLIP: Adapting CLIP for Universal Visual Anomaly Detectioncited by 28
- 2025PAG: Multi-Turn Reinforced LLM Self-Correction with Policy as Generative Verifiercited by 28
- 2023CoTFormer: A Chain-of-Thought Driven Architecture with Budget-Adaptive Computation Cost at Inferencecited by 27
- 2024MANTA: A Large-Scale Multi-View and Visual-Text Anomaly Detection Dataset for Tiny Objectscited by 26
- 2024Multi-Sensor Object Anomaly Detection: Unifying Appearance, Geometry, and Internal Propertiescited by 26
- 2025LogicAD: Explainable Anomaly Detection via VLM-based Text Feature Extractioncited by 24
- 2025OmniAD: Detect and Understand Industrial Anomaly via Multimodal Reasoningcited by 18
- 2025IAD-R1: Reinforcing Consistent Reasoning in Industrial Anomaly Detectioncited by 17
- 2025LR-IAD:Mask-Free Industrial Anomaly Detection with Logical Reasoningcited by 9
Built on by
Citing papers per year: 2026: 1.
1 paper.
Similar, not cited
Close in wording, no citation either way
- 2025Can Multimodal Large Language Models be Guided to Improve Industrial Anomaly Detection?
- 2026EAGLE: Expert-Augmented Attention Guidance for Tuning-Free Industrial Anomaly Detection in Multimodal Large Language Models
- 2025AD-FM: Multimodal LLMs for Anomaly Detection via Multi-Stage Reasoning and Fine-Grained Reward Optimization
- 2026MMR-AD: A Large-Scale Multimodal Dataset for Benchmarking General Anomaly Detection with Multimodal Large Language Models
- 2025PB-IAD: Utilizing multimodal foundation models for semantic industrial anomaly detection in dynamic manufacturing environments
- 2025Detect, Classify, Act: Categorizing Industrial Anomalies with Multi-Modal Large Language Models
- 2026A Structured Benchmark for Text-Guided Anomaly Detection: When Language Stops Conditioning the Decision
- 2026ADOPD: Reference-Privileged On-Policy Distillation for MLLM-Based Industrial Anomaly Detection