# CNC: Cross-modal Normality Constraint for Unsupervised Multi-class Anomaly Detection

arXiv: https://arxiv.org/abs/2501.00346 · HTML: https://mlsanity.com/paper/2501.00346/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 31 Dec 2024.

## Built on by (most cited first, 5 of 5)

- [Normal-Abnormal Guided Generalist Anomaly Detection](https://mlsanity.com/paper/2510.00495/) (2025)
- [AnomalyMoE: Towards a Language-free Generalist Model for Unified Visual Anomaly Detection](https://mlsanity.com/paper/2508.06203/) (2025)
- [Teacher Encoder-Student Decoder Denoising Guided Segmentation Network for Anomaly Detection](https://mlsanity.com/paper/2501.12104/) (2025)
- [MIRAD - A comprehensive real-world robust anomaly detection dataset for Mass Individualization](https://mlsanity.com/paper/2510.16370/) (2025)
- [Exploring Multimodal Prompts For Unsupervised Continuous Anomaly Detection](https://mlsanity.com/paper/2603.21562/) (2026)

## Abstract (arXiv)

Existing unsupervised distillation-based methods rely on the differences between encoded and decoded features to locate abnormal regions in test images. However, the decoder trained only on normal samples still reconstructs abnormal patch features well, degrading performance. This issue is particularly pronounced in unsupervised multi-class anomaly detection tasks. We attribute this behavior to over-generalization(OG) of decoder: the significantly increasing diversity of patch patterns in multi-class training enhances the model generalization on normal patches, but also inadvertently broadens its generalization to abnormal patches. To mitigate OG, we propose a novel approach that leverages class-agnostic learnable prompts to capture common textual normality across various visual patterns, and then apply them to guide the decoded features towards a normal textual representation, suppressing over-generalization of the decoder on abnormal patterns. To further improve performance, we also introduce a gated mixture-of-experts module to specialize in handling diverse patch patterns and reduce mutual interference between them in multi-class training. Our method achieves competitive performance on the MVTec AD and VisA datasets, demonstrating its effectiveness.