ICLAD: In-Context Learning for Unified Tabular Anomaly Detection Across Supervision Regimes

TL;DR

ICLAD introduces a unified in-context learning model for tabular anomaly detection that effectively operates across various supervision regimes without retraining, demonstrating state-of-the-art results on numerous datasets.

Contribution

The paper presents ICLAD, a meta-learned foundation model that generalizes anomaly detection across datasets and supervision levels using in-context learning, a novel approach in this domain.

Findings

Achieves state-of-the-art performance on 57 datasets.

Effectively handles multiple supervision regimes.

Operates without updating model weights at inference.

Abstract

Anomaly detection on tabular data is commonly studied under three supervision regimes, including one-class settings that assume access to anomaly-free training samples, fully unsupervised settings with unlabeled and potentially contaminated training data, and semi-supervised settings with limited anomaly labels. Existing deep learning approaches typically train dataset-specific models under the assumption of a single supervision regime, which limits their ability to leverage shared structures across anomaly detection tasks and to adapt to different supervision levels. We propose ICLAD, an in-context learning foundation model for tabular anomaly detection that generalizes across both datasets and supervision regimes. ICLAD is trained via meta-learning on synthetic tabular anomaly detection tasks, and at inference time, the model assigns anomaly scores by conditioning on the training set without updating model weights. Comprehensive experiments on 57 tabular datasets from ADBench show that our method achieves state-of-the-art performance across three supervision regimes, establishing a unified framework for tabular anomaly detection.