mlsanityData to 14 Oct 2026

RangeAD: Fast On-Model Anomaly Detection

arXiv 2603.17795 ↗First posted 18 Mar 2026cs.LG cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

In practice, machine learning methods commonly require anomaly detection (AD) to filter inputs or detect distributional shifts. Typically, this is implemented by running a separate AD model alongside the primary model. However, this separation ignores the fact that the primary model already encodes substantial information about the target distribution. In this paper, we introduce On-Model AD, a setting for anomaly detection that explicitly leverages access to a related machine learning model. Within this setting, we propose RangeAD, an algorithm that utilizes neuron-wise output ranges derived from the primary model. RangeAD achieves superior performance even on high-dimensional tasks while incurring substantially lower inference costs. Our results demonstrate the potential of the On-Model AD setting as a practical framework for efficient anomaly detection.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 18 of its 52 references that are in the map. First posted on arXiv 18 Mar 2026.

Builds on

18 of 52 references.

References · 18

  1. 2013Visualizing and Understanding Convolutional Networkscited by 2,290
  2. 2021Swin Transformer V2: Scaling Up Capacity and Resolutioncited by 1,228
  3. 2016Temporal Convolutional Networks for Action Segmentation and Detectioncited by 508
  4. 2020Underspecification Presents Challenges for Credibility in Modern Machine Learningcited by 496
  5. 2020A Unifying Review of Deep and Shallow Anomaly Detectioncited by 425
  6. 2022ADBench: Anomaly Detection Benchmarkcited by 232
  7. 2019PyOD: A Python Toolbox for Scalable Outlier Detectioncited by 222
  8. 2020Classification-Based Anomaly Detection for General Datacited by 211
  9. 2022ECOD: Unsupervised Outlier Detection Using Empirical Cumulative Distribution Functionscited by 153
  10. 2020COPOD: Copula-Based Outlier Detectioncited by 146
  11. 2023On Diffusion Modeling for Anomaly Detectioncited by 72
  12. 2021"Garbage In, Garbage Out" Revisited: What Do Machine Learning Application Papers Report About Human-Labeled Training Data?cited by 24
6 more
  1. 2023Weakly Supervised Detection of Hallucinations in LLM Activationscited by 20
  2. 2021Online-compatible Unsupervised Non-resonant Anomaly Detectioncited by 15
  3. 2023Meta-survey on outlier and anomaly detectioncited by 13
  4. 2026MacrOData: New Benchmarks of Thousands of Datasets for Tabular Outlier Detectioncited by 5
  5. 2025We Need to Rethink Benchmarking in Anomaly Detectioncited by 4
  6. 2022Fault Prognosis in Particle Accelerator Power Electronics Using Ensemble Learningcited by 2

Built on by

Not cited yet.