Skip to content
conftrace_
2026 EMNLP EMNLP 2026

A Geometry-Aware Depth Function for Robust Unsupervised Textual Anomaly Detection

AbstractTextual anomaly detection (AD) aims to identify samples that deviate from the underlying distribution of normal data, without requiring explicit supervision on anomalies. Existing literature has focused almost exclusively on the semi-supervised setting, assuming access to clean training datasets composed solely of normal samples. However, this assumption rarely holds in real-world scenarios, as datasets used to infer the normal distribution are frequently contaminated by anomalies. In this work, we first highlight the sensitivity of conventional anomaly detection methods to contamination, before demonstrating the relevance of data depth functions for robust textual anomaly detection. Building on recent findings regarding the geometry of language model embeddings, further supported by our empirical observations, we introduce the geodesic depth to better capture the structure of contextual embedding spaces. Our results show that the geodesic depth consistently outperforms commonly used anomaly detection methods once contamination exceeds 10% in the training data, with substantially larger improvements observed in the most challenging scenarios.

More from EMNLP 2026

All EMNLP 2026 papers →