Towards One-for-All Anomaly Detection for Tabular Data

TL;DR

OFA-TAD is a novel one-for-all anomaly detection framework for tabular data that generalizes across diverse datasets with a single training phase, leveraging neighbor-distance patterns and multi-view representations.

Contribution

The paper introduces OFA-TAD, a generalist framework that enables one-time training on multiple datasets and effective generalization to unseen domains for tabular anomaly detection.

Findings

OFA-TAD outperforms existing methods on 34 datasets from 14 domains.

It demonstrates strong cross-domain generalization capabilities.

The approach achieves superior anomaly detection accuracy under the one-for-all setting.

Abstract

Tabular anomaly detection (TAD) aims to identify samples that deviate from the majority in tabular data and is critical in many real-world applications. However, existing methods follow a ``one model for one dataset (OFO)'' paradigm, which relies on dataset-specific training and thus incurs high computational cost and yields limited generalization to unseen domains. To address these limitations, we propose OFA-TAD, a generalist one-for-all (OFA) TAD framework that only requires one-time training on multiple source datasets and can generalize to unseen datasets from diverse domains on-the-fly. To realize one-for-all tabular anomaly detection, OFA-TAD extracts neighbor-distance patterns as transferable cues, and introduces multi-view neighbor-distance representations from multiple transformation-induced metric spaces to mitigate the transformation sensitivity of distance profiles. To adaptively combine multi-view distance evidence, a Mixture-of-Experts (MoE) scoring network is employed for view-specific anomaly scoring and entropy-regularized gated fusion, with a multi-strategy anomaly synthesis mechanism to support training under the one-class constraint. Extensive experiments on 34 datasets from 14 domains demonstrate that OFA-TAD achieves superior anomaly detection performance and strong cross-domain generalizability under the strict OFA setting.