Overview

Anomaly localization studies how to find the specific pixels, regions, frames, time intervals, or variables responsible for an unusual observation, so that detection can support inspection and diagnosis. Common approaches learn normal patterns with reconstruction models, pretrained visual features, or memory banks, then produce anomaly maps; weakly supervised video models and vision-language systems use image- or video-level labels, prompts, or reasoning to ground their predictions. In time-series and infrastructure settings, forecasting, causal graphs, and counterfactual analysis help locate affected variables or likely root causes. A central challenge is making localization reliable under domain shift and limited labels, while evaluating spatial or temporal precision separately from image- or video-level detection scores.

Latest papers

CardsList

158 papers11 papers in the last four weeks, down 31% on the four weeks before. 0.1% of all new papers.

  1. RAID: Retrieval-Augmented Anomaly Detection

    Feb 23, 2026Mingxiu Cai, Zhe Zhang, Gaochang Wu +2Unsupervised Anomaly DetectionAnomaly Detection

  2. AnomaMind: Coarse-to-Fine Agentic Diagnosis via Hybrid Inference for Time Series Anomaly Detection

    Feb 14, 2026Xiaoyu Tao, Yuchong Wu, Mingyue Cheng +2Time-Series Anomaly DetectionLLM-Based Time Series Anomaly Detection

  3. Root Cause Analysis of Measurement and Mechanistic Anomalies

    Jan 30, 2026Hendrik Suhr, David Kaltenpoth, Jilles VreekenRoot Cause AnalysisAnomaly Localization

  4. Explainable Visual Anomaly Detection via Concept Bottleneck Models

    Nov 25, 2025Arianna Stropeni, Valentina Zaccaria, Francesco Borsatti +3Interpretable Anomaly DetectionAnomaly Localization

  5. Structured Temporal Causality for Interpretable Multivariate Time Series Anomaly Detection

    Oct 18, 2025Dongchan Cho, Jiho Han, Keumyeong Kang +3Time-Series Anomaly DetectionUnsupervised Anomaly Detection

  6. Diverse Normal Prototypes-Guided Contrastive Reconstruction for Medical Anomaly Detection

    Aug 27, 2025Luhu Li, Bin Liu, Bowen Lin +3Prototype LearningMedical Image Anomaly Detection

  7. 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