# A Masked Reverse Knowledge Distillation Method Incorporating Global and Local Information for Image Anomaly Detection

arXiv: https://arxiv.org/abs/2512.15326 · HTML: https://mlsanity.com/paper/2512.15326/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Detection and dense prediction distillation, one of 12 directions in Knowledge distillation.
- First posted on arXiv 17 Dec 2025.

## Placed in

- Knowledge distillation / [Detection and dense prediction distillation](https://mlsanity.com/distill/detection-and-dense-prediction-distillation/)

## Built on by (most cited first, 3 of 3)

- [VTFusion: A Vision-Text Multimodal Fusion Network for Few-Shot Anomaly Detection](https://mlsanity.com/paper/2601.16381/) (2026)
- [A Feature Shuffling and Restoration Strategy for Universal Unsupervised Anomaly Detection](https://mlsanity.com/paper/2603.22861/) (2026)
- [Bidirectional Adaptive Transformers for Multimodal Anomaly Detection](https://mlsanity.com/paper/2610.14598/) (2026)

## Abstract (arXiv)

Knowledge distillation is an effective image anomaly detection and localization scheme. However, a major drawback of this scheme is its tendency to overly generalize, primarily due to the similarities between input and supervisory signals. In order to address this issue, this paper introduces a novel technique called masked reverse knowledge distillation (MRKD). By employing image-level masking (ILM) and feature-level masking (FLM), MRKD transforms the task of image reconstruction into image restoration. Specifically, ILM helps to capture global information by differentiating input signals from supervisory signals. On the other hand, FLM incorporates synthetic feature-level anomalies to ensure that the learned representations contain sufficient local information. With these two strategies, MRKD is endowed with stronger image context capture capacity and is less likely to be overgeneralized. Experiments on the widely-used MVTec anomaly detection dataset demonstrate that MRKD achieves impressive performance: image-level 98.9% AU-ROC, pixel-level 98.4% AU-ROC, and 95.3% AU-PRO. In addition, extensive ablation experiments have validated the superiority of MRKD in mitigating the overgeneralization problem.