Industrial Anomaly Detection

Overview

Industrial anomaly detection aims to identify defects or unusual behavior when examples of failures are scarce and normal operation can vary across products, machines, and conditions. Visual methods often learn normal patterns with reconstruction models, patch-memory methods such as PatchCore, or pretrained vision features; synthetic defects and vision-language models are increasingly used to extend detection to new defect types. For sensor and audio data, common approaches include forecasting and reconstruction, autoencoders, and graph models that capture relationships among signals, with benchmarks such as MVTec AD, VisA, SWaT, and WADI. A central challenge is moving from strong results on curated benchmarks to reliable deployment under shifts, contaminated training data, tight false-alarm limits, and edge-device constraints.

Latest papers

CardsList

276 papers24 papers in the last four weeks, down 23% on the four weeks before. 0.2% of all new papers.

  1. Digital Twin-Driven Communication-Efficient Federated Anomaly Detection for Industrial IoT

    Jan 5, 2026Mohammed Ayalew Belay, Adil Rasheed, Pierluigi Salvo RossiCommunication-Efficient Distributed TrainingIndustrial Anomaly Detection

  2. Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation

    Dec 10, 2025Nadeem Nazer, Hongkuan Zhou, Lavdim Halilaj +2Industrial Anomaly DetectionMultimodal Anomaly Detection

  3. AD-FM: Multimodal LLMs for Anomaly Detection via Multi-Stage Reasoning and Fine-Grained Reward Optimization

    Aug 6, 2025Jingyi Liao, Yongyi Su, Rong-Cheng Tu +6VLM AdaptationIndustrial Anomaly Detection

  4. ISP-AD: A Large-Scale Real-World Dataset for Advancing Industrial Anomaly Detection with Synthetic and Real Defects

    Mar 6, 2025Paul J. Krassnig, Dieter P. GruberIndustrial Anomaly DetectionIndustrial Inspection

  5. Tuned Reverse Distillation: Enhancing Multimodal Industrial Anomaly Detection with Crossmodal Tuners

    Dec 12, 2024Xinyue Liu, Jianyuan Wang, Biao Leng +1Cross-Modal Knowledge DistillationIndustrial Anomaly Detection