Results show that embedding-space geometry contains predictive information about anomaly detection performance while also highlighting the representation-dependent nature of inlier-only performance estimates.
Abstract
Anomaly detection systems are often trained using normal data alone, while model selection and evaluation typically require labeled anomalies. We study whether anomaly detection performance can be predicted without access to anomalous data. For kNN-based detectors, we derive a lower bound on the area under the ROC curve (AUC) that relates detection performance to the separation between inlier and outlier scores and to their respective variances. Under a local scaling model, we use this bound to characterize how density variation, intrinsic-dimensional heterogeneity, and cross-domain mismatch contribute to score variability. We then investigate anomaly-free model selection and show that inlier score variance alone does not reliably predict performance across different representations. To address this limitation, we introduce simple pseudo-anomaly probes that provide a reference for estimating relative score separation. Experiments on the DCASE 2022-2025 benchmarks, spanning four embedding models and 208 candidate systems, show that pseudo-anomaly-based estimators substantially improve anomaly-free model selection. In particular, diverse pseudo-anomalies enable anomaly-free model selection to outperform conventional development-set selection under domain shift. These results show that embedding-space geometry contains predictive information about anomaly detection performance while also highlighting the representation-dependent nature of inlier-only performance estimates.
This work uses the AUC bound as a differentiable, anomaly-free objective for directly optimizing continuous parameters of anomaly detection systems and demonstrates this framework by optimizing ensemble weights and introducing a learnable score-rescaling mechanism that adapts pseudo-anomaly scores, enabling optimizatio...
Log Reconstruction and Distance (LoRD), a lightweight post-hoc calibration framework for reliable log anomaly detection, is proposed and demonstrates that LoRD consistently improves confidence reliability and substantially reduces overconfident anomaly-related errors without sacrificing anomaly detection performance.
LLM-Detector is proposed, a framework that utilizes the in-context learning capacity of LLMs for structured, prompt-conditioned scoring synthesis, enabling LLMs to derive anomaly detection logic from structured normal-state knowledge.
Tu Nguyen, Dang Nguyen, T. D. Le et al.· 0 citations
The original single-step Decision score used by TCCM is sensitive to contamination, whereas trajectory-based Deviation and Reconstruction scores provide more stable anomaly signals, and Forest-Flow becomes competitive with, and in some cases outperforms, TCCM.
Philip Konz, Tejaswini Medi, Margret Keuper· 0 citations
This work investigates whether automated interpretability can augment visual anomaly detectors by identifying and intervening on different components of this signal, and demonstrates that individual feature interventions across all 43 categories show that the MLLM labels are aligned in aggregate with how features diffe...
Antonio De Santis, Arsenio Leo, Marco Brambilla· arXiv.org· 0 citations
A representation-enhanced AD model with prototype learning and neighborhood rough set (ADPLN) is proposed, which outperforms the state-of-the-art methods and designs a sample-prototype correlation objective function to promote the learning of discriminative representations of normals and mitigate the negative impacts o...
Jiao-Long Chen, Ye Liu, Yu-Hua Qian et al.· IEEE Transactions on Cyberne...· 0 citations
Related blog posts
MIT News · Artificial Intelligence· news.mit.eduNov 17, 2026
By studying the computational foundations of language learning, Associate Professor Jacob Andreas builds intelligent systems that can communicate effectively with humans.
Microsoft Research Blog· microsoft.comNov 16, 2026
AI coding agents are increasingly help developers fix bugs and build features largely on their own, but every interaction with their models consumes tokens. In each session, an agent sends instructions, tool descriptions, and session history to its model. Most efforts to reduce this cost focus on individual tasks. Yet developers return to the same […] The post TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient appeared first on Microsoft Research.
MIT News · Artificial Intelligence· news.mit.eduNov 13, 2026