WMoE-CLIP: Wavelet-Enhanced Mixture-of-Experts Prompt Learning for Zero-Shot Anomaly Detection

2026-03-06

AI Summary

This work proposes a novel framework for zero-shot anomaly detection that overcomes the limitations of fixed text prompts and single spatial-domain features, which struggle to capture complex semantics and subtle anomalies. The approach integrates multi-frequency visual analysis with semantic adaptability by employing a variational autoencoder to model global semantics and dynamically refine CLIP text embeddings. Concurrently, wavelet decomposition is leveraged to extract multi-scale frequency-domain features. A semantic-aware mixture-of-experts module is further introduced to enable fine-grained cross-modal alignment. Notably, this is the first method to combine wavelet-based multi-frequency analysis with mixture-of-experts prompt learning. Extensive experiments on 14 industrial and medical datasets demonstrate significant performance gains over existing approaches, highlighting its superior generalization capability.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multimodal Learning
• Data Mining & Knowledge Management: Anomaly/Outlier Detection

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding

Abstract

Vision-language models have recently shown strong generalization in zero-shot anomaly detection (ZSAD), enabling the detection of unseen anomalies without task-specific supervision. However, existing approaches typically rely on fixed textual prompts, which struggle to capture complex semantics, and focus solely on spatial-domain features, limiting their ability to detect subtle anomalies. To address these challenges, we propose a wavelet-enhanced mixture-of-experts prompt learning method for ZSAD. Specifically, a variational autoencoder is employed to model global semantic representations and integrate them into prompts to enhance adaptability to diverse anomaly patterns. Wavelet decomposition extracts multi-frequency image features that dynamically refine textual embeddings through cross-modal interactions. Furthermore, a semantic-aware mixture-of-experts module is introduced to aggregate contextual information. Extensive experiments on 14 industrial and medical datasets demonstrate the effectiveness of the proposed method.

Problem

• zero-shot anomaly detection
• textual prompts
• spatial-domain features
• subtle anomalies
• vision-language models

Innovation

• Wavelet Decomposition
• Mixture-of-Experts
• Prompt Learning
• Zero-Shot Anomaly Detection
• Cross-modal Interaction