Anomaly Detection Foundation Models

A collection of papers on anomaly detection (tabular data/time series/image/video/graph/text/log) with the large language model, large visual model, and graph foundation model.

We will continue to update this list with the latest resources. If you find any missed resources (paper/code) or errors, please feel free to open an issue or make a pull request.

Tabular data

• [Li2024] Anomaly Detection of Tabular Data Using LLMs in IJCAI-W, 2024. [paper][code]

• [Tsai2025] AnoLLM: Large Language Models for Tabular Anomaly Detection in ICLR, 2025. [paper][code]

• [Chen2025] PyOD 2: A Python Library for Outlier Detection with LLM-powered Model Selection in Arxiv, 2025. [paper][code]

Time series

• [Liu2024] Large Language Model Guided Knowledge Distillation for Time Series Anomaly Detection in IJCAI, 2024. [paper][code]

• [Liu2024] Large Language Models can Deliver Accurate and Interpretable Time Series Anomaly Detection in Arxiv, 2024. [paper][code]

• [Alnegheimish2024] Large language models can be zero-shot anomaly detectors for time series? in DSAA, 2024. [paper][code]

• [Shentu2025] Towards a General Time Series Anomaly Detector with Adaptive Bottlenecks and Dual Adversarial Decoders in ICLR, 2025. [paper][code]

• [Zhou2025] Can LLMs Understand Time Series Anomalies? in ICLR, 2025. [paper][code]

• [Xu2025] Can Multimodal LLMs Perform Time Series Anomaly Detection? in Arxiv, 2025. [paper][code]

• [Chen2025] Synergizing Large Language Models and Task-specific Models for Time Series Anomaly Detection in Arxiv, 2025. [paper][code]

• [Wu2025] Uncertainty-Aware Fine-Tuning for Time Series Anomaly Detection in Openreview, 2025. [paper][code]

• [Yang2025] Refining Time Series Anomaly Detectors using Large Language Models in Arxiv, 2025. [paper][code]

• [Wang2025] Pre-training Enhanced Transformer for multivariate time series anomaly detection in Information Fusion, 2025. [paper][code]

• [Maru2025] RATFM: Retrieval-augmented Time Series Foundation Model for Anomaly Detection in Arxiv, 2025. [paper][code]

• [Garcia2025] Towards Foundation Auto-Encoders for Time-Series Anomaly Detection in KDDW, 2025. [paper][code]

• [Park2025] When Will It Fail?: Anomaly to Prompt for Forecasting Future Anomalies in Time Series in ICML, 2025. [paper][code]

Image

• [Jeong2023] WinCLIP: Zero-/Few-Shot Anomaly Classification and Segmentation in CVPR, 2023. [paper][code]

• [Chen2024] CLIP-AD: A Language-Guided Staged Dual-Path Model for Zero-shot Anomaly Detection in Arxiv, 2024. [paper][code]

• [Gu2024] AnomalyGPT: Detecting Industrial Anomalies Using Large Vision-Language Models in AAAI, 2024. [paper][code]

• [Zhou2024] AnomalyCLIP: Object-agnostic Prompt Learning for Zero-shot Anomaly Detection in ICLR, 2024. [paper][code]

• [Zhu2024] Toward Generalist Anomaly Detection via In-context Residual Learning with Few-shot Sample Prompts in CVPR, 2024. [paper][code]

• [Li2024] PromptAD: Learning Prompts with only Normal Samples for Few-Shot Anomaly Detection in CVPR, 2024. [paper][code]

• [Xu2024] Customizing Visual-Language Foundation Models for Multi-modal Anomaly Detection and Reasoning in Arxiv, 2024. [paper][code]

• [Zhu2024] Do LLMs Understand Visual Anomalies? Uncovering LLM's Capabilities in Zero-shot Anomaly Detection in MM, 2024. [paper][code]

• [Li2024] One-to-Normal: Anomaly Personalization for Few-shot Anomaly Detection in NeurIPS, 2024. [paper][code]

• [Zuo2024] CLIP3D-AD: Extending CLIP for 3D Few-Shot Anomaly Detection with Multi-View Images Generation [paper][code]

• [Zhu2025] Fine-grained Abnormality Prompt Learning for Zero-shot Anomaly Detection in Arxiv, 2025. [paper][code]

• [Tao2025] Kernel-Aware Graph Prompt Learning for Few-Shot Anomaly Detection in AAAI, 2025. [paper][code]

• [Xu2025] Towards Zero-Shot Anomaly Detection and Reasoning with Multimodal Large Language Models in CVPR, 2025. [paper][code]

• [Qu2025] Bayesian Prompt Flow Learning for Zero-Shot Anomaly Detection in CVPR, 2025. [paper][code]

• [Ma2025] AA-CLIP: Enhancing Zero-Shot Anomaly Detection via Anomaly-Aware CLIP in CVPR, 2025. [paper][code]

• [Zhang2025] Towards Training-free Anomaly Detection with Vision and Language Foundation Models in CVPR, 2025. [paper][code]

• [Gu2025] UniVAD: A Training-free Unified Model for Few-shot Visual Anomaly Detection in CVPR, 2025. [paper][code]

• [Yun2025] Language-Assisted Feature Transformation for Anomaly Detection in ICLR, 2025. [paper][code]

• [Lv2025] One-for-All Few-Shot Anomaly Detection via Instance-Induced Prompt Learning in ICLR, 2025. [paper][code]

• [Jiang2025] MMAD: A Comprehensive Benchmark for Multimodal Large Language Models in Industrial Anomaly Detection in ICLR, 2025. [paper][code]

• [Guo2025] Dinomaly: The Less Is More Philosophy in Multi-Class Unsupervised Anomaly Detection in CVPR, 2025. [paper][code]

• [Li2025] One-for-More: Continual Diffusion Model for Anomaly DetectionOne-for-More: Continual Diffusion Model for Anomaly Detection in CVPR, 2025. [paper][code]

• [Zeng2025] Towards Efficient and General-Purpose Few-Shot Misclassification Detection for Vision-Language Models in Arxiv, 2025. [paper][code]

• [Luo2025] Exploring Intrinsic Normal Prototypes within a Single Image for Universal Anomaly Detection in CVPR, 2025. [paper][code]

• [Sun2025] Anomaly Anything: Promptable Unseen Visual Anomaly Generation in CVPR, 2025. [paper][code]

• [Sadikaj2025] MultiADS: Defect-aware Supervision for Multi-type Anomaly Detection and Segmentation in Zero-Shot Learning in Arxiv, 2025. [paper][code]

• [Zhao2025] AnomalyHybrid: A Domain-agnostic Generative Framework for General Anomaly Detection in CVPRW, 2025. [paper][code]

• [Kim2025] GenCLIP: Generalizing CLIP Prompts for Zero-shot Anomaly Detection in Arxiv, 2025. [paper][code]

• [Chao2025] AnomalyR1: A GRPO-based End-to-end MLLM for Industrial Anomaly Detection in Arxiv, 2025. [paper][code]

• [Hu2025] ReplayCAD: Generative Diffusion Replay for Continual Anomaly Detection in IJCAI, 2025. [paper][code]

• [Gao2025] AdaptCLIP: Adapting CLIP for Universal Visual Anomaly Detection in Arxiv, 2025. [paper][code]

• [Yang2025] ViP-CLIP: Visual-Perception Prompting with Unified Alignment for Zero-Shot Anomaly Detection in Arxiv, 2025. [paper][code]

• [Shiri2025] MadCLIP: Few-shot Medical Anomaly Detection with CLIP in MICCAI, 2025. [paper][code]

Video

• [Wu2023] VadCLIP: Adapting Vision-Language Models for Weakly Supervised Video Anomaly Detection in Arxiv, 2024. [paper][code]

• [Yang2024] Follow the Rules: Reasoning for Video Anomaly Detection with Large Language Models in Arxiv, 2024. [paper][code]

• [Zhang2024] Holmes-VAD: Towards Unbiased and Explainable Video Anomaly Detection via Multi-modal LLM in Arxiv, 2024. [paper][code]

• [Wu2024] VadCLIP: Adapting Vision-Language Models for Weakly Supervised Video Anomaly Detection in AAAI, 2024. [paper][code]

• [Zanella2024] Harnessing Large Language Models for Training-free Video Anomaly Detection in CVPR, 2024. [paper][code]

• [Yang2024] Text Prompt with Normality Guidance for Weakly Supervised Video Anomaly Detection in CVPR, 2024. [paper][code]

• [Cho2024] Towards Multi-Domain Learning for Generalizable Video Anomaly Detection in NeurIPS, 2024. [paper][code]

• [Lv2024] Video Anomaly Detection and Explanation via Large Language Models in ICCV, 2024. [paper][code]

• [Wu2024] Weakly Supervised Video Anomaly Detection and Localization with Spatio-Temporal Prompts in ACM MM, 2024. [paper][code]

• [Ye2025] VERA: Explainable Video Anomaly Detection via Verbalized Learning of Vision-Language Models in CVPR, 2025. [paper][code]

• [Wu2025] AVadCLIP: Audio-Visual Collaboration for Robust Video Anomaly Detection in Arxiv, 2025. [paper][code]

• [Yang2025] AssistPDA: An Online Video Surveillance Assistant for Video Anomaly Prediction in Arxiv, 2025. [paper][code]

• [Ding2025] SlowFastVAD: Video Anomaly Detection via Integrating Simple Detector and RAG-Enhanced Vision-Language Model in Arxiv, 2025. [paper][code]

• [Shao2025] EventVAD: Training-Free Event-Aware Video Anomaly Detection in Arxiv, 2025. [paper][code]

• [Huang2025] Vad-R1: Towards Video Anomaly Reasoning via Perception-to-Cognition Chain-of-Thought in Arxiv, 2025. [paper][code]

• [Lee2025] Flashback: Memory-Driven Zero-shot, Real-time Video Anomaly Detection in Arxiv, 2025. [paper][code]

• [Orlova2025] Simplifying Traffic Anomaly Detection with Video Foundation Models in Arxiv, 2025. [paper][code]

• [Zang2025] SAGE: A Visual Language Model for Anomaly Detection via Fact Enhancement and Entropy-aware Alignment in MM, 2025. [paper][code]

• [Mu2025] NexViTAD: Few-shot Unsupervised Cross-Domain Defect Detection via Vision Foundation Models and Multi-Task Learning in Arxiv, 2025. [paper][code]

Graph

• [Liu2024] ARC: A Generalist Graph Anomaly Detector with In-Context Learning in NeurIPS, 2024. [paper][code]

• [Lin2024] UniGAD: Unifying Multi-level Graph Anomaly Detection in NeurIPS, 2024. [paper][code]

• [Niu2024] Zero-shot Generalist Graph Anomaly Detection with Unified Neighborhood Prompts in IJCAI, 2024. [paper][code]

• [Qiao2025] AnomalyGFM: Graph Foundation Model for Zero/Few-shot Anomaly Detection in KDD, 2025. [paper][code]

• [Xu2025] GLIP-OOD: Zero-Shot Graph OOD Detection with Foundation Model in Arxiv, 2025. [paper][code]

• [Zheng2025] DP-DGAD: A Generalist Dynamic Graph Anomaly Detector with Dynamic Prototypes in Aexiv, 2025. [paper][code]

Text

• [Yang2024] AD-LLM: Benchmarking Large Language Models for Anomaly Detection in Arxiv, 2024. [paper][code]

• [Yang2025] Fraud-R1: A Multi-Round Benchmark for Assessing the Robustness of LLM Against Augmented Fraud and Phishing Inducements in Arxiv, 2025. [paper][code]

Log

• [Karlsen2023] Benchmarking Large Language Models for Log Analysis, Security, and Interpretation Arxiv, 2023. [paper][code]

• [Qi2023] LogGPT: Exploring ChatGPT for Log-Based Anomaly Detection in Arixv, 2024. [paper][code]

• [Almodovar2024] LogFiT: Log Anomaly Detection Using Fine-Tuned Language Models in TNSM, 2024. [paper][code]

• [Guan2025] LogLLM: Log-based Anomaly Detection Using Large Language Models in Arxiv, 2025. [paper][code]

• [Lim2025] Adapting Large Language Models for Parameter-Efficient Log Anomaly Detection in PAKDD, 2025. [paper][code]

• [Song2025] Confront Insider Threat: Precise Anomaly Detection in Behavior Logs Based on LLM Fine-Tuning in Coling, 2025. [paper][code]

• [Yang2025] LogLLaMA: Transformer-based log anomaly detection with LLaMA in Arxiv, 2025. [paper][code]

• [Liu2024] Interpretable Online Log Analysis Using Large Language Models with Prompt Strategies in ICPC [paper][code]

• [Xu2025] OpenRCA: Can Large Language Models Locate the Root Cause of Software Failures in ICLR, 2025. [paper][code]

Related Survey

• [Ren2025] Foundation Models for Anomaly Detection: Vision and Challenges in Arxiv, 2025. [paper][code]

• [Su2025] Large Language Models for Forecasting and Anomaly Detection: A Systematic Literature Review in Arxiv, 2025. [paper][code]

• [Xu2025] Large Language Models for Anomaly and Out-of-Distribution Detection: A Survey in Arxiv, 2025. [paper][code]