When Normality Shifts: Risk-Aware Test-Time Adaptation for Unsupervised Tabular Anomaly Detection

TL;DR

This paper introduces RTTAD, a risk-aware test-time adaptation method for unsupervised tabular anomaly detection that improves robustness to normality shifts by combining dual-task training and selective test-time updates.

Contribution

It proposes a holistic two-stage approach with dual-task learning during training and risk-aware selective adaptation during testing, enhancing anomaly detection under normality shifts.

Findings

Achieves state-of-the-art detection performance on 15 tabular datasets.

Effectively mitigates normality shift issues through risk-aware test-time adaptation.

Refines embedding distributions with a k-NN contrastive objective.

Abstract

Unsupervised tabular anomaly detection methods typically learn feature patterns from normal samples during training and subsequently identify samples that deviate from these patterns as anomalies during testing. However, in practical scenarios, the limited scale and diversity of training data often lead to an incomplete characterization of normal patterns. While test-time adaptation offers a remedy, its isolated focus on test-time optimization ignores the critical synergy with training-phase learning. Furthermore, indiscriminate adaptation to unlabeled test data inevitably triggers anomaly contamination, preventing the model from fully realizing its discriminative capability between normal and anomalous samples. To address these issues, we propose RTTAD, a Risk-aware Test-time adaptation method for unsupervised Tabular Anomaly Detection. RTTAD holistically tackles normality shifts via a synergistic two-stage mechanism. During training, collaborative dual-task learning captures multi-level representations to establish a robust normal prior. During testing, a Test-Time Contrastive Learning (TTCL) module explicitly accounts for adaptation risk by selectively updating the model using high-confidence pseudo-normal samples while constraining anomalous ones. Additionally, TTCL incorporates a k-nearest neighbor-based contrastive objective to refine embedding distributions, thereby further enhancing the model's discriminative capacity. Extensive experiments on 15 tabular datasets demonstrate that RTTAD achieves state-of-the-art overall detection performance.