Diffuse to Detect: Generative Diffusion Models for Unsupervised IC Anomaly Detection
On the map: Diffusion models → Flow matching, time series and science
Abstract (arXiv)
Latent defect screening is challenged by extremely low failure rates, high-dimensional test data, and absence of labeled anomalies. We propose the first unsupervised anomaly detection framework incorporating a Diffusion Transformer. Raw test measurements are first compressed by an autoencoder, then reshaped into a structured token sequence enriched with sinusoidal and per-device wafer-position embeddings. Anomaly scores are derived from the noise-prediction error over mid-range diffusion timesteps, enabling fast wafer-scale screening without any labeled defects or manual feature engineering. Our approach achieves state-of-the-art performance on industrial 16nm IC test data under extreme class imbalance, offering interpretable failure localization through latent-space reconstruction residuals.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Placed in Flow matching, time series and science, one of 14 directions in Diffusion models. Builds on 20 of its 42 references that are in the map. First posted on arXiv 26 May 2026.
Builds on
20 of 42 references.
Closest ancestors · 5
Same direction, newest first
- 2023TransFusion -- A Transparency-Based Diffusion Model for Anomaly Detection16 Nov
- 2023ImDiffusion: Imputed Diffusion Models for Multivariate Time Series Anomaly Detection3 Jul
- 2023On Diffusion Modeling for Anomaly Detection29 May
- 2023Anomaly Detection with Conditioned Denoising Diffusion Models25 May
- 2022TabDDPM: Modelling Tabular Data with Diffusion Models30 Sep
Other references · 15
- 2017Attention Is All You Needcited by 59,932
- 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
- 2020Denoising Diffusion Probabilistic Modelscited by 20,851
- 2022Scalable Diffusion Models with Transformerscited by 5,931
- 2021Improved Denoising Diffusion Probabilistic Modelscited by 3,736
- 2015Variational Inference with Normalizing Flowscited by 2,161
- 2018GANomaly: Semi-Supervised Anomaly Detection via Adversarial Trainingcited by 409
- 2019PyOD: A Python Toolbox for Scalable Outlier Detectioncited by 221
- 2020Classification-Based Anomaly Detection for General Datacited by 210
- 2022ECOD: Unsupervised Outlier Detection Using Empirical Cumulative Distribution Functionscited by 153
- 2020COPOD: Copula-Based Outlier Detectioncited by 143
- 2020DROCC: Deep Robust One-Class Classificationcited by 121
3 more
- 2023Fascinating Supervisory Signals and Where to Find Them: Deep Anomaly Detection with Scale Learningcited by 25
- 2023Unsupervised 3D out-of-distribution detection with latent diffusion modelscited by 22
- 2023Generative Pre-Training of Time-Series Data for Unsupervised Fault Detection in Semiconductor Manufacturingcited by 1
Built on by
Citing papers per year: 2026: 1.
1 paper.
Similar, not cited
Close in wording, no citation either way
- 2026Efficient Patch-Based Anomaly Detection Fused with Diffusion Driven Generative Modeling for Semiconductor Wafer Bin Map Open Set Anomaly Detection
- 2025ExDD: Explicit Dual Distribution Learning for Surface Defect Detection via Diffusion Synthesis
- 2024Leveraging Latent Diffusion Models for Training-Free In-Distribution Data Augmentation for Surface Defect Detection
- 2024Addressing Class Imbalance and Data Limitations in Advanced Node Semiconductor Defect Inspection: A Generative Approach for SEM Images
- 2023DiffusionAD: Norm-guided One-step Denoising Diffusion for Anomaly Detection
- 2024A Novel Approach to Industrial Defect Generation through Blended Latent Diffusion Model with Online Adaptation
- 2024FabGPT: An Efficient Large Multimodal Model for Complex Wafer Defect Knowledge Queries
- 2026Industrial Surface Defect Detection via Diffusion Generation and Asymmetric Student-Teacher Network