AD-AGENT: A Multi-agent Framework for End-to-end Anomaly Detection

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yang, Tiankai, Liu, Junjun, Siu, Wingchun, Wang, Jiahang, Qian, Zhuangzhuang, Song, Chanjuan, Cheng, Cheng, Hu, Xiyang, Zhao, Yue
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908369360519168
author Yang, Tiankai
Liu, Junjun
Siu, Wingchun
Wang, Jiahang
Qian, Zhuangzhuang
Song, Chanjuan
Cheng, Cheng
Hu, Xiyang
Zhao, Yue
author_facet Yang, Tiankai
Liu, Junjun
Siu, Wingchun
Wang, Jiahang
Qian, Zhuangzhuang
Song, Chanjuan
Cheng, Cheng
Hu, Xiyang
Zhao, Yue
contents Anomaly detection (AD) is essential in areas such as fraud detection, network monitoring, and scientific research. However, the diversity of data modalities and the increasing number of specialized AD libraries pose challenges for non-expert users who lack in-depth library-specific knowledge and advanced programming skills. To tackle this, we present AD-AGENT, an LLM-driven multi-agent framework that turns natural-language instructions into fully executable AD pipelines. AD-AGENT coordinates specialized agents for intent parsing, data preparation, library and model selection, documentation mining, and iterative code generation and debugging. Using a shared short-term workspace and a long-term cache, the agents integrate popular AD libraries like PyOD, PyGOD, and TSLib into a unified workflow. Experiments demonstrate that AD-AGENT produces reliable scripts and recommends competitive models across libraries. The system is open-sourced to support further research and practical applications in AD.
format Preprint
id arxiv_https___arxiv_org_abs_2505_12594
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle AD-AGENT: A Multi-agent Framework for End-to-end Anomaly Detection
Yang, Tiankai
Liu, Junjun
Siu, Wingchun
Wang, Jiahang
Qian, Zhuangzhuang
Song, Chanjuan
Cheng, Cheng
Hu, Xiyang
Zhao, Yue
Computation and Language
Artificial Intelligence
Anomaly detection (AD) is essential in areas such as fraud detection, network monitoring, and scientific research. However, the diversity of data modalities and the increasing number of specialized AD libraries pose challenges for non-expert users who lack in-depth library-specific knowledge and advanced programming skills. To tackle this, we present AD-AGENT, an LLM-driven multi-agent framework that turns natural-language instructions into fully executable AD pipelines. AD-AGENT coordinates specialized agents for intent parsing, data preparation, library and model selection, documentation mining, and iterative code generation and debugging. Using a shared short-term workspace and a long-term cache, the agents integrate popular AD libraries like PyOD, PyGOD, and TSLib into a unified workflow. Experiments demonstrate that AD-AGENT produces reliable scripts and recommends competitive models across libraries. The system is open-sourced to support further research and practical applications in AD.
title AD-AGENT: A Multi-agent Framework for End-to-end Anomaly Detection
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2505.12594