A Unified Model for Multi-Class Anomaly Detection
Open accessThe DOI matches its Crossref record; 51 of the 51 references it cites were found in a registry or an index; 14 of the 27 sources its claims cite were checked for support.
Abstract
Despite the rapid advance of unsupervised anomaly detection, existing methods require to train separate models for different objects. In this work, we present UniAD that accomplishes anomaly detection for multiple classes with a unified framework. Under such a challenging setting, popular reconstruction networks may fall into an 'identical shortcut', where both normal and anomalous samples can be well recovered, and hence fail to spot outliers. To tackle this obstacle, we make three improvements, including a layer-wise query decoder to help model the multi-class distribution. When learning a unified model for 15 categories in MVTec-AD, we surpass the second competitor on the tasks of both anomaly detection (from 88.1% to 96.5%) and anomaly localization (from 89.5% to 96.8%).
Overview
CoChat Summary
TL;DR
UniAD is a unified feature-reconstruction framework for multi-class anomaly detection that avoids the “identical shortcut” via a layer-wise query decoder, neighbor masked attention, and feature jittering. On MVTec-AD’s 15-category unified setting it reaches 96.5% AUROC for anomaly detection and 96.8% for localization, substantially outperforming prior methods.
Key findings
- On MVTec-AD with one unified model for 15 categories, UniAD achieves 96.5% AUROC for anomaly detection versus 88.1% for the second competitor DRAEM, and 96.8% for anomaly localization versus 89.5% for the second competitor PaDiM.
- On CIFAR-10 under the proposed many-versus-many unified setting, UniAD reaches 87.2% mean AUROC, outperforming FCDD+OE by 8.3% without auxiliary outlier exposure; baselines US and FCDD fall to 55.9% and 55.8% mean AUROC.
- Ablations show the query embedding is critical: without query embedding, detection/localization AUROC is 69.5%/79.4%; with one query embedding it rises to 87.6%/92.8%; with layer-wise query it reaches 95.0%/96.5%; combining layer-wise query, neighbor masked attention, and feature jittering gives 96.5%/96.8%.
- Transformer-based competitors under the unified setting perform far worse: InTra 65.3%/70.6%, VT-ADL 55.4%/64.4%, AnoViT 69.6%/68.4%, while a UniAD baseline with one query embedding already reaches 87.6%/92.8%.
Methods
- UniAD reconstructs features from a fixed ImageNet-pretrained EfficientNet-b4 backbone: stage-1 to stage-4 features are resized, concatenated into a 272-channel map, tokenized, and linearly projected to 256 channels; the model uses a Neighbor Masked Encoder and Layer-wise Query Decoder, with neighbor masked attention masking each token’s self and neighbors, and feature jittering adding Gaussian noise to input tokens.
- Training on MVTec-AD uses 224×224 images, 14×14 feature size, 1000 epochs on 8 NVIDIA Tesla V100 16GB GPUs, batch size 64, AdamW with weight decay 1×10⁻⁴, initial learning rate 1×10⁻⁴ dropped by 0.1 after 800 epochs, 4 encoder and 4 decoder layers, neighbor size 7×7, jittering scale 20, jittering probability 1, and 5 random seeds; CIFAR-10 uses stage-1 to stage-5 features forming a 720-channel map and batch size 128.
- The objective is MSE feature reconstruction loss; anomaly localization scores are the L2 norm of reconstruction differences, upsampled to image size, and image-level anomaly detection takes the maximum of the average-pooled score map; evaluation uses AUROC under the unified case without fine-tuning.
Limitations
- Stated in the paper: the method does not use category labels, and incorporating labels into the unified model should be further studied; practical normal samples are not as consistent as MVTec-AD, though UniAD handles all 15 MVTec-AD categories; anomaly detection for video surveillance may infringe personal privacy.
- Evident from the text: evaluation is limited to MVTec-AD and CIFAR-10, with no real-world video-surveillance experiments despite the privacy discussion; the unified setting assumes class labels are inaccessible at both training and inference, so the method cannot exploit category information that might improve multi-class modeling.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
8 Valid · 5 Not valid · 6 Not checked · 95 Own results and interpretations
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Despite the rapid advance of unsupervised anomaly detection, existing methods require to train separate models for different objects.”
- Also stated in: § 1 Introduction · p. 1 · plain
“existing methods [6, 11, 26, 28, 49, 50, 53] propose to train separate models for different classes of objects”
What the sources say
- Could not verifycites [6] [6] P. Bergmann, M. Fauser, D. Sattlegger, and C. Steger. Uninformed students: Student-teacher anomaly detection with discriminative latent embeddings. In IEEE Conf. Comput. Vis. Pattern Recog. , 2020. - sourceExperimental — accuracy not yet measured
Our copy repeatedly characterises the shortcomings of existing unsupervised anomaly detection methods, but never in terms of needing a separate model per object class: the Introduction (p. 1) blames 'shallow machine learning pipelines that require a dimensionality reduction of the used feature space' and 'heavy training data subsampling', Related Work (p. 2) and Sec. 2.1 (p. 2) blame from-scratch generative models, per-pixel reconstruction errors and patch-sampled shallow classifiers, and the Introduction (p. 1) notes that one-/multi-class methods 'assume that anomalies manifest themselves in the form of images of an entirely different class'. Searches for 'separate model per object/class', 'single model for all categories' and per-category training returned no passage making the attributed point; conversely the paper itself trains on 'a single class of anomaly-free images' and fits models per MVTec AD category (Sec. 4.2, p. 7), so nothing in our copy states or contradicts the claim that existing methods require object-specific models. (LLM, conf=0.70)
- Partly supportedcites [11] [11] T. Defard, A. Setkov, A. Loesch, and R. Audigier. PaDim: A patch distribution modeling framework for anomaly detection and localization. In Int. Conf. Pattern Recog. , 2021. - sourceExperimental — accuracy not yet measured
The Experimental setups section (p. 4) says the paper's VAE baseline is trained on each MVTec AD class, supporting class-specific training as an example. But the Scalability gain section (p. 6) says SPADE requires no training, so the paper does not support the broader claim that existing methods generally require separate models for different objects. (LLM, conf=0.85)
- Not checked — no full text of the sourcecites [26] [26] C.-L. Li, K. Sohn, J. Yoon, and T. Pfister. CutPaste: Self-supervised learning for anomaly detection and localization. In IEEE Conf. Comput. Vis. Pattern Recog. , 2021. - source
- Partly supportedcites [28] [28] P. Liznerski, L. Ruff, R. A. Vandermeulen, B. J. Franks, M. Kloft, and K.-R. Müller. Explainable deep one-class classification. In Int. Conf. Learn. Represent. , 2021. - sourceExperimental — accuracy not yet measured
Section 4.1 (p. 5) describes a common one-vs-rest anomaly-detection benchmark that treats each class in turn as nominal, consistent with training class-specific models. It does not establish that existing unsupervised anomaly-detection methods generally require a separate model for every object class. (LLM, conf=0.84)
- Not checked — no full text of the sourcecites [49] [49] J. Yi and S. Yoon. Patch SVDD: Patch-level SVDD for anomaly detection and segmentation. In Asian Conf. Comput. Vis. , 2020. - source
- Partly supportedcites [50] [50] Z. You, K. Yang, W. Luo, L. Cui, Y. Zheng, and X. Le. ADTR: Anomaly detection transformer with feature reconstruction. arXiv preprint arXiv:2209.01816 , 2022. - sourceExperimental — accuracy not yet measured
The Dataset section (p. 6) describes a one-class CIFAR-10 setup in which the training set of one class is used for training, supporting class-specific training in that setting. The paper does not establish that existing unsupervised anomaly detection methods generally require separate models for different objects. (LLM, conf=0.75)
- Not checked — no full text of the sourcecites [53] [53] V. Zavrtanik, M. Kristan, and D. Skoˇ caj. DRAEM-A discriminatively trained reconstruction embedding for surface anomaly detection. In Int. Conf. Comput. Vis. , 2021. - source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Also stated in: § 1 Introduction · p. 1 · plain
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“EfficientNet performs obviously better than ResNet, especially in anomaly detection.”
What the sources say
- Not checked — no full text of the sourcecites [20] [20] K. He, X. Zhang, S. Ren, and J. Sun. Deep residual learning for image recognition. In IEEE Conf. Comput. Vis. Pattern Recog. , 2016. - source
- Partly supportedcites [41] [41] M. Tan and Q. Le. EfficientNet: Rethinking model scaling for convolutional neural networks. In Int. Conf. Mach. Learn. , 2019. - sourceExperimental — accuracy not yet measured
The Introduction (p. 2) reports higher ImageNet accuracy for EfficientNet-B4 than ResNet-50, and the transfer-learning discussion (p. 8) reports better accuracy–parameter performance than ResNet. The paper does not report an anomaly-detection comparison, so it does not support the claim that EfficientNet performs especially better in that task. (LLM, conf=0.90)
boostedstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Every source the paper cites for this claim was checked and supports it.
In the paper
“Adding only one query embedding to the first decoder layer ( i.e. , vanilla transformer [43]) promotes the performance dramatically by 18.1% and 13.4% in anomaly detection and localization, respectively.”
What the sources say
- Supportedcites [43] [43] A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, Ł. Kaiser, and I. Polosukhin. Attention is all you need. In Adv. Neural Inform. Process. Syst. , 2017. - source
The citation is attached to “vanilla transformer,” identifying the architecture used as a baseline, rather than to the reported performance gains. The Introduction (p. 2) proposes the Transformer, and Section 3.1 (p. 3) describes its decoder layers. The 18.1% and 13.4% results are not attributed to this source by that citation. (LLM, conf=0.92)
boostedstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The sources with a verdict support this claim; the rest weren’t checked yet, have no full text we could read, or their verdicts aren’t rated yet.
In the paper
“Inspired by Denoising Auto-Encoder (DAE) [3, 44], we add perturbations to feature tokens, guiding the model to learn knowledge of normal samples by the denoising task.”
What the sources say
- Supportedcites [3] [3] Y. Bengio, L. Yao, G. Alain, and P. Vincent. Generalized denoising auto-encoders as generative models. In Adv. Neural Inform. Process. Syst. , 2013. - source
The Introduction describes corrupting observed inputs and training a denoising auto-encoder to reconstruct them. The Conclusion says denoising training implicitly estimates the underlying data-generating process. The citing authors present perturbing feature tokens as their own DAE-inspired implementation, not as a procedure used in this source. (LLM, conf=0.90)
- Not checked — no full text of the sourcecites [44] [44] P. Vincent, H. Larochelle, Y. Bengio, and P.-A. Manzagol. Extracting and composing robust features with denoising autoencoders. In Int. Conf. Mach. Learn. , 2008. - source
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Rests on 2 assumptions not shown here.
- Supported
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“NMA together with FJ provides a quite strong performance (96.2% for detection and 96.6% for localization).”
boostedstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
72 more claims
Create a free account to see all 114 claims and every source verdict.
- It may state two separate claims as one. 25
- Probably not a substantial claim. 14
- The quote only partly states this claim. 2
Citations
Cites
The publisher hasn’t deposited this paper’s reference list.
Research works printed but not linked: 4
- [10][10] L. Deecke, L. Ruff, R. A. Vandermeulen, and H. Bilen. Transfer-based semantic anomaly detection. In Int. Conf. Mach. Learn. , 2021. -
- [22][22] L. Ilya and H. Frank. Decoupled weight decay regularization. In Int. Conf. Learn. Represent. , 2019. -
- [24][24] A. Krizhevsky. Learning multiple layers of features from tiny images. Master's thesis, University of Tront , 2009. -
- [36][36] L. Ruff, R. Vandermeulen, N. Goernitz, L. Deecke, S. A. Siddiqui, A. Binder, E. Müller, and M. Kloft. Deep one-class classification. In Int. Conf. Mach. Learn. , 2018. -
Cited by
Items 101–200 of 377
ADFormer: Generalizable Few-Shot Anomaly Detection With Dual CNN-Transformer Architecture
Zhu, B., Gu, Z., Zhu, G., Chen, Y., Tang, M., & Wang, J. (2024) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2024.3522624 · Cited by 17
ROADS: Robust Prompt-Driven Multi-Class Anomaly Detection Under Domain Shift
Kashiani, H., Talemi, N. A., & Afghah, F. (2025) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv61041.2025.00768 · Cited by 12
Unilaterally Aggregated Contrastive Learning with Hierarchical Augmentation for Anomaly Detection
Wang, G., Wang, Y., Qin, J., Zhang, D., Bao, X., & Huang, D. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00634 · Cited by 12
GAN-Based Anomaly Detection Tailored for Classifiers
Králik, Ľ., Kontšek, M., Škvarek, O., & Klimo, M. (2024) · Mathematics · doi:10.3390/math12101439 · Cited by 8
LogicAL: Towards logical anomaly synthesis for unsupervised anomaly localization
Zhao, Y. (2024) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw63382.2024.00406 · Cited by 13
Dissolving is Amplifying: Towards Fine-Grained Anomaly Detection
Shi, J., Zhang, P., Zhang, N., Ghazzai, H., & Wonka, P. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-73202-7_22 · Cited by 11
Few-Shot Anomaly Detection with Adversarial Loss for Robust Feature RepresentationsPreprint
Lee, J. Y., Lee, W., Choi, J., Lee, Y., & Yoon, Y. S. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2312.03005 · Cited by 10
Anomaly Detection in Medical Images Using Encoder-Attention-2Decoders Reconstruction
Tang, P., Yan, X., Hu, X., Wu, K., Lasser, T., & Shi, K. (2025) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2025.3563482 · Cited by 12
Automated Defect Report Generation for Enhanced Industrial Quality Control
Xie, J., Zhou, Z., Wu, Z., Zhang, X., Wang, J., Cai, Y., & Li, Q. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i17.29900 · Cited by 11
Segmenting Remote Sensing Anomalies at Instance Level via Anomaly Map-Guided Adaptation
Li, J., Zhong, Y., Zhao, H., Gao, Z. Y., & Wang, X. (2024) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2024.3439491 · Cited by 9
Steering knuckle surface defect detection and segmentation based on reverse residual distillation
Zhou, S., Cheng, S., Zhang, D., Wang, Z., Zhang, S., Zhu, Y., & Wang, H. (2024) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2024.109161 · Cited by 8
A simple and reliable semi-supervised anomaly detection network for detecting crack in stamped parts
Dong, X., Zhang, C., Liu, S., & Wang, D. (2025) · Computers in Industry · doi:10.1016/j.compind.2025.104301 · Cited by 7
Unified Anomaly Detection Methods on Edge Device Using Knowledge Distillation and Quantization
Jena, S., Pulkit, A., Singh, K., Banerjee, A., Joshi, S., Ganesh, A., Singh, D., & Bhavsar, A. V. (2025) · Lecture notes in computer science · doi:10.1007/978-3-031-97822-7_5 · Cited by 9
Unveiling Multi-View Anomaly Detection: Intra-view Decoupling and Inter-view Fusion
Mao, K., Lian, Y., Wang, Y., Liu, M., Zheng, N., & Wei, P. (2025) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v39i12.33349 · Cited by 11
LScAD: A Large–Small Model Collaboration Framework for Unsupervised Industrial Anomaly Detection
Qu, S., Tao, X., Gong, X., Qu, Z., Prasad, M., Shen, F., Zhang, Z., & Ding, G. (2025) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2025.3588927 · Cited by 8
Jiang, X., Chen, Y., Nie, Q., Liu, J., Liu, Y., Wang, C., & Zheng, F. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2403.14213 · Cited by 15
MistSense: Versatile Online Detection of Procedural and Execution Mistakes
Patsch, C., Wu, Y., Zakour, M., Salihu, D., & Steinbach, E. G. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.01348 · Cited by 8
Enhancing random surface anomaly detection in real-world using a four-stage one-class approach
Li, P., Wu, G., Zhou, Y., & Leng, J. (2025) · Pattern Recognition Letters · doi:10.1016/j.patrec.2025.05.002 · Cited by 6
A memory-tree driven network for multi-view fusion anomaly detection
Liu, X., Wu, C., Zhang, H., & Wang, L. (2025) · Pattern Recognition · doi:10.1016/j.patcog.2025.112106 · Cited by 5
Multimodal feature cooperative refinement for few-shot anomaly detection
Xu, L., Han, D., Li, G., Zhou, M., Wan, J., & Li, M. (2025) · Advanced Engineering Informatics · doi:10.1016/j.aei.2025.103792 · Cited by 5
Multiclass unsupervised anomaly detection with directional feature perturbation
Li, Z., Yang, L., Liu, J., & Huang, X. (2025) · Measurement Science and Technology · doi:10.1088/1361-6501/ae035c · Cited by 5
Liu, Y., Zhang, C., Dong, X., & Yang, Y. (2025) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2025.112299 · Cited by 7
CFFDist: Cross-Scale Feature Fusion Distillation Network for Industrial Anomaly Localization
Zhi, H., Qin, H., Zhang, L., Guo, J., & Song, B. (2024) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2024.3502773 · Cited by 6
Li, S., Li, Z., Wang, W., Wang, W., Zheng, L., & Yu, L. (2025) · Neurocomputing · doi:10.1016/j.neucom.2025.130085 · Cited by 6
Memory association guided unsupervised anomaly detection with adaptive 3D attention
Liu, X., Wu, C., Zhang, H., & Wang, L. (2025) · Information Fusion · doi:10.1016/j.inffus.2025.103379 · Cited by 7
Robust anomaly detection in industrial images by blending global–local features
Pei, M., Liu, N., & Xia, S. (2024) · Expert Systems · doi:10.1111/exsy.13624 · Cited by 5
BiGSeT: Binary Mask-Guided Separation Training for DNN-based Hyperspectral Anomaly DetectionPreprint
Liu, H., Su, X., Shen, X., Chen, L., & Zhou, X. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2307.07428 · Cited by 10
XMatchAD: A Cross-Modal Matching Perspective on Reconstruction-Based Anomaly Detection
Cai, M., Zhang, Z., Wu, G., & Chai, T. (2026) · IEEE Transactions on Image Processing · doi:10.1109/tip.2026.3718437 · Cited by 4
UMLLA-AD: Mamba-Driven Adaptive Feature Selection for Industrial Anomaly Detection
Fang, T., Wang, J., Ye, M., & Huang, Y. (2025) · International Conference on Multimedia Retrieval · doi:10.1145/3731715.3733458 · Cited by 4
ADPretrain: Advancing Industrial Anomaly Detection via Anomaly Representation Pretraining
Yao, X., Luo, Y., Qian, Z., & Zhang, C. (2025) · neural information processing systems · doi:10.52202/085713-4101 · Cited by 7
Enhanced KPI Anomaly Detection: An Unsupervised Hybrid Model with Dynamic Threshold
Wang, Y., Chen, T., Tang, Y., & Huang, L. (2024) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp48485.2024.10446434 · Cited by 4
Wang, X., He, J., Huang, F.-S., Shen, S., & Liu, Z. (2024) · Journal of Energy Storage · doi:10.1016/j.est.2024.113743 · Cited by 5
Silalahi, S., Ahmad, T., Studiawan, H., Anthi, E., & Williams, L. (2025) · IEEE Access · doi:10.1109/access.2025.3580056 · Cited by 4
Wu, W., & Pan, C. (2025) · Frontiers in Public Health · doi:10.3389/fpubh.2025.1624345 · Cited by 4
Industrial anomaly detection using mask attention-based discriminative reconstruction networks
Chen, R., Liu, L., Tian, J. R., Ai, Y., & Zhang, F. (2025) · Computers & Electrical Engineering · doi:10.1016/j.compeleceng.2025.110784 · Cited by 4
Su, X., Shen, X., Liu, H., Chen, L., Vivone, G., & Zhou, X. (2025) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2025.3580751 · Cited by 4
Xi, Z., Gao, Y., Li, X., & Gao, L. (2026) · Pattern Recognition · doi:10.1016/j.patcog.2026.113728 · Cited by 3
Unsupervised Tablet Defect Detection Method Based on Diffusion Model
Zhang, M., Liu, W., He, L., & Wang, D. (2025) · Sensors · doi:10.3390/s25175254 · Cited by 3
Shan, P., Liang, T., He, D., Pan, G., Zhi, M., & Zhao, Y. (2025) · Measurement · doi:10.1016/j.measurement.2025.119155 · Cited by 4
Unsupervised anomaly detection with a stacked transformer diffusion reconstruction framework
Du, M., Xu, H., Shang, F., Wang, Z., Hu, Y., & Xie, L. (2026) · Expert Systems with Applications · doi:10.1016/j.eswa.2026.131764 · Cited by 4
Pixel-Level Contrastive Pretrainer for Industrial Image Representation
Zhu, B., Chen, Y., Tang, M., & Wang, J. (2024) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2024.3353860 · Cited by 6
Efficient textile anomaly detection via memory guided distillation network
Yang, J., Wang, H., Song, Z., Guo, F., & Yue, H. (2024) · Journal of Intelligent Manufacturing · doi:10.1007/s10845-024-02445-9 · Cited by 3
Multi-level Feature Adaptation and Embeddings Alignment for Zero-shot Anomaly Detection
Liu, Y., Deng, H., & Zhao, G. L. (2025) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49660.2025.10889575 · Cited by 3
Insulator Surface Defect Detection Method Based on Graph Feature Diffusion Distillation
Li, S., Na, Z., Yang, G., Hou, Y., & Zhang, X. (2025) · Journal of Imaging · doi:10.3390/jimaging11060190 · Cited by 3
Image anomaly detection with a unified transformer model guided by dual-feature
Wang, Y., Jing, J., & Zhang, X. (2025) · Journal of Industrial Information Integration · doi:10.1016/j.jii.2025.100892 · Cited by 6
Latent Diffusion Based Multi-class Anomaly Detection
Chenxing, W., & Tavakkoli, A. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-47969-4_38 · Cited by 3
PixOOD: Pixel-Level Out-of-Distribution Detection
Vojíř, T., Sochman, J., & Matas, J. (2026) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2026.3688748 · Cited by 7
Cluster-Wide Task Slowdown Detection in Cloud System
Chen, F., Zhang, Y., Fan, L., Liang, Y., Pang, G., Wen, Q., & Deng, S. (2024) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3637528.3671936 · Cited by 3
Detecting Both Seen and Unseen Anomalies in Time Series
Liu, C., He, S., Li, S., Shi, Z., & Meng, W. (2025) · ACM Transactions on Knowledge Discovery from Data · doi:10.1145/3717071 · Cited by 5
Uniformaly: Towards Task-Agnostic Unified Framework for Visual Anomaly DetectionPreprint
Lee, Y., Lim, H., Jang, S., & Yoon, H. (2023) · SSRN Electronic Journal · doi:10.2139/ssrn.4586132 · Cited by 4
Dual-Path Frequency Discriminators for Few-Shot Anomaly DetectionPreprint
Bai, Y., Zhang, J., Dong, Y., Cao, Y., & Tian, G. (2024) · SSRN Electronic Journal · doi:10.2139/ssrn.4862099 · Cited by 3
Zhong, Y., Guo, B., Sun, Z., Deng, L., Xu, X., & Dian, S. (2026) · Computers in Industry · doi:10.1016/j.compind.2026.104467 · Cited by 4
Wang, X., Chai, L., & Chen, J. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.02134 · Cited by 2
Dual-Masked and Discriminative Reconstruction for Unified Vision Anomaly Detection
Gao, B.-B. (2026) · IEEE Transactions on Image Processing · doi:10.1109/tip.2026.3687095 · Cited by 3
A Distillation Network using Improved ConvLSTM for Video Anomaly Detection
Li, J., Zhou, C., Liu, S., & Lu, X. (2024) · International Conference on Computing, Machine Learning, and Data Science (CMLDS) · doi:10.1145/3661725.3661779 · Cited by 2
Self-citationDisentangled Knowledge Distillation for Unified Multi-Class Anomaly Detection
Jang, J.-Y., Lee, H., & Lee, Y. (2024) · Proceedings - International Conference on Image Processing · doi:10.1109/icip51287.2024.10647795 · Cited by 2
IGSPAD: Inverting 3D Gaussian Splatting for Pose-agnostic Anomaly Detection
Jiang, B., Xie, Y., Li, J., Li, N., Chen, B., & Xia, S. (2024) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3664647.3681619 · Cited by 4
A unified multi‐class anomaly detection model based on reverse distillation
Fu, M., & Fu, Z. (2024) · IET Image Processing · doi:10.1049/ipr2.13309 · Cited by 2
Adaptive Continual Learning for Defect Detection in Dynamic Textile Manufacturing
Harith, S., Rajapakse, C., Mahanama, T. V., & Asanka, D. (2024) · International Conference on Advancements in Computing (ICAC) · doi:10.1109/icac64487.2024.10850925 · Cited by 2
Introducing Fast Fourier Convolutions into Anomaly Detection
Zhao, Z., & Zhou, J. (2025) · Sensors · doi:10.3390/s25165196 · Cited by 2
Kang, B., Zhong, Y., Deng, L., Wang, M., & Zhang, J. (2025) · Neurocomputing · doi:10.1016/j.neucom.2025.131824 · Cited by 4
PP-Former: Exploring purified intrinsic normal prototypes for industrial anomaly detection
Wei, H., Zhao, L., Li, R., Hu, G., Zhang, M., Ouyang, B., Ruan, C., Tan, J., & Xu, Y. (2025) · Knowledge-Based Systems · doi:10.1016/j.knosys.2025.114759 · Cited by 5
MPFR: Memory prompt feature reconstruction for continual anomaly detection and segmentation
Chen, Y., Tao, X., Chen, B., Wang, J., Zhang, Y., Chun, P., Li, X., & Zhou, X. (2026) · Pattern Recognition · doi:10.1016/j.patcog.2025.112946 · Cited by 5
S3AD: Efficient industrial anomaly detection via selection, space, and scale
Peng, Z., Dong, C., Liu, B., Li, Z., Zeng, K., & Li, J. (2026) · Pattern Recognition · doi:10.1016/j.patcog.2026.113868 · Cited by 2
FG-VAD: A fast and general framework for few-shot visual anomaly detection
Xu, Q., Wu, Y., Song, K., & Xiong, H. (2026) · Neurocomputing · doi:10.1016/j.neucom.2026.133876 · Cited by 2
ResAD++: Towards Class Agnostic Anomaly Detection via Residual Feature Learning
Yao, X., Shi, C., Zhao, M., Zhai, G., & Zhang, C. L. (2026) · International Journal of Computer Vision · doi:10.1007/s11263-026-02885-1 · Cited by 2
Language-Guided Mixture of Spectral Experts for Cross-Scenario Hyperspectral Anomaly Detection
Fu, Y., Ma, P., Xu, Y., Ng, M. K., & Ren, P. (2026) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2026.3719083 · Cited by 4
A Decoder-Free Reconstruction Method for Semi-Supervised Rail Surface Defect Detection
Liu, C., Shi, Z., He, S., Tang, S., & Yang, Q. (2024) · IEEE Transactions on Industrial Cyber-Physical Systems · doi:10.1109/ticps.2024.3456758 · Cited by 2
DualAD: Exploring Coupled Dual-Branch Networks for Multi-Class Unsupervised Anomaly Detection
He, S., Chen, Y., Wang, L., Huang, W., Xu, R., & Qian, Y. (2025) · Electronics · doi:10.3390/electronics14030594 · Cited by 4
MemFlow-AD: An anomaly detection and localization model based on memory module and normalizing flow
Xie, X., Li, Z., Xiong, S., Liu, Z., & Cai, T. (2025) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2025.104454 · Cited by 3
Unsupervised Document Image Tampering Localization via Anomaly Detection
Li, Y., Zhang, Y., Yin, F., & Huang, L.-L. (2025) · Lecture notes in computer science · doi:10.1007/978-981-96-6606-5_7 · Cited by 2
Arundhathi, V., & Shine, L. (2025) · International Conference on Control, Communication and Computing (ICCC) · doi:10.1109/iccc64910.2025.11077266 · Cited by 2
Towards efficient pixel labeling for industrial anomaly detection and localization
Wu, J., Li, H., Wu, L. Y., Chen, H., Liu, D., & Jin, H. (2026) · Pattern Recognition Letters · doi:10.1016/j.patrec.2026.06.020 · Cited by 3
Nearest Memory Augmented Feature Reconstruction for Unified Anomaly Detection
Wu, H., Wang, C., Jian, Z., Lai, Y., Song, L., & Yang, F. (2023) · Communications in computer and information science · doi:10.1007/978-981-99-8148-9_28 · Cited by 3
Friend or Foe? Harnessing Controllable Overfitting for Anomaly DetectionPreprint
Long, Q., Zhu, B., Chen, Y., Tang, M., & Wang, J. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2412.00560 · Cited by 2
GA-CLIP: Dual-branch CLIP for generalizable and anomaly-aware zero-shot anomaly detection
Guo, W., Wan, Y., Yuan, S., Chang, L., Wu, Y., & Dou, Z. (2026) · Knowledge-Based Systems · doi:10.1016/j.knosys.2026.116078 · Cited by 1
A Unified Wafer Anomaly Detection and Segmentation
Lin, S.-C., Hwang, J., & Lai, S. (2026) · ASMC proceedings · doi:10.1109/asmc69324.2026.11551192 · Cited by 1
ERCAD: An embedding replay method for continual anomaly detection and segmentation
Deng, Z., Yang, G., Tu, B., Liu, Y., & Man, J. (2026) · Pattern Recognition · doi:10.1016/j.patcog.2026.114507 · Cited by 1
PromptMAD: Cross-Modal Prompting for Multi-Class Visual Anomaly Localization
McCain, D., Kashiani, H., & Afghah, F. (2026) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp55912.2026.11464635 · Cited by 1
Unified Multimodal Industrial Anomaly Detection via Few Normal Samples
Jiang, S., Ma, Y., Zhou, J., Wang, Y., & Liu, M. (2026) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2026.3681017 · Cited by 2
ArcAD: Anomaly-Rectified Calibration for Cold-Start Supervised Anomaly Detection
Han, N., Fan, L., Guo, J., Cao, Y., Su, X., Cao, F., Di, D., & Su, T. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37092-1_26 · Cited by 2
IMMoE: Incomplete Multi-view Anomaly Detection via Mixture of View Experts Fusion
Hu, L. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37041-9_3 · Cited by 1
InDeed: Interpretable Image Deep Decomposition With Grounded Generalizability
Wang, S., Gao, S., Wu, F., & Zhuang, X. (2026) · IEEE Transactions on Image Processing · doi:10.1109/tip.2026.3718452 · Cited by 1
Text-Guided Multimodal Unified Industrial Anomaly DetectionPreprint
Li, Z., Ye, S., Yu, Z., Xie, W., & Shen, L. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.7243962 · Cited by 1
A comprehensive survey on continual anomaly detection
Angiulli, F., Luca, F. D., & Fassetti, F. (2026) · Neurocomputing · doi:10.1016/j.neucom.2026.135118 · Cited by 1
Pixel Logo Attack: Embedding Attacks as Logo-Like Pixels
Zhu, J., Zhao, H., Yu, H. L., & Liu, J. (2024) · Genetic and Evolutionary Computation Conference (GECCO) · doi:10.1145/3638529.3654231 · Cited by 1
Multi-Class Hypersphere Anomaly Detection Based on Edge Outlier Exposure Set and Margin
Gao, M., Liu, X., Xu, D., & Yang, G. (2024) · Mathematics · doi:10.3390/math12152340 · Cited by 3
Time Series Anomaly Detection via Reconstruction-Limited Probability Estimation
Feng, Y., & Niu, Y. (2025) · Lecture notes in computer science · doi:10.1007/978-981-97-8702-9_30 · Cited by 2
Learning Decision Boundaries for Multidimensional Anomaly Detection
Wang, X., Duan, L., Guan, L., Xu, J., & He, C. (2025) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2025.3540602 · Cited by 2
Wang, H., Shen, K., & Wang, F. (2025) · Circuits Systems and Signal Processing · doi:10.1007/s00034-025-03064-2 · Cited by 1
Tang, N., Chen, Q., Meng, Y., Lei, D., Jiang, L., Qin, Y., Huang, X., Tang, F., Huang, S., Lan, Q., et al. (2025) · Frontiers in Bioengineering and Biotechnology · doi:10.3389/fbioe.2025.1576513 · Cited by 3
Transformer-Guided Serial Knowledge Distillation for High-Precision Anomaly Detection
Wang, D., & Wang, B. (2025) · IEEE Access · doi:10.1109/access.2025.3584892 · Cited by 1
Using Deep Learning to Detect Stitching Artefacts in Drone-Based Orthomosaics of Orchards
Gordon, D., Whales, J., Smith, A., & Marais, P. (2025) · Communications in computer and information science · doi:10.1007/978-3-031-96262-2_16 · Cited by 1
Masked variational transformer for complex clutter modeling and target detection
Shi, L., Liang, X., Chen, W., Liu, Y., Ding, T., Qin, K., Chen, B., & Liu, H. (2025) · Signal Processing · doi:10.1016/j.sigpro.2025.110236 · Cited by 2
An industrial defect detection method based on mixed noise synthesis
Ke, A., Luo, J., Yu, Y., Li, L., & Cai, B. (2025) · Computers in Industry · doi:10.1016/j.compind.2025.104388 · Cited by 1
Entropy-increasing linear attention for multi-class unsupervised anomaly detection
Liu, T., Gao, H., Tan, Y., Li, J., & Zhao, J. (2026) · Pattern Recognition · doi:10.1016/j.patcog.2026.113089 · Cited by 2
Wei, H., Zhao, L., Li, R., Hu, G., Zhang, M., Ouyang, B., Ruan, C., Tan, J., & Xu, Y. (2026) · Results in Engineering · doi:10.1016/j.rineng.2026.109285 · Cited by 1
CRAD-HOPE: brain-inspired nested learning framework for few-shot anomaly detection
Cao, W., Cao, J., & Wu, W. (2026) · Journal of King Saud University - Computer and Information Sciences · doi:10.1007/s44443-026-00577-x · Cited by 2
Dynamic expandable framework for incremental anomaly detection
Tan, Y., Gao, H., Liu, T., & Wen, X. (2026) · Neural Networks · doi:10.1016/j.neunet.2026.108949 · Cited by 1
GroundingAnomaly: Spatially-Grounded Diffusion for Few-Shot Anomaly Synthesis
Liu, Y., Chen, H., Zhao, P., Bao, Y., Tian, Y., Zhang, J., Chen, H., Zhi, Z., Liu, Y., Li, Y., et al. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-36987-1_6 · Cited by 2
Metadata
- Authors
- You, Z., Cui, L., Shen, Y., Yang, K., Lu, X., Zheng, Y., & Le, X.
- ORCID
- Lu, Xin
- Venue
- Advances in Neural Information Processing Systems 35
- Published
- 2022
- Pages
- 4571-4584
- ISSN
- 1049-5258 (ISSN-L)
- OpenAlex
- W7133227337
- Cited by
- 303
- References
- 0
- Citations by year
- 2026: 115 · 2025: 76 · 2024: 87 · 2023: 25
- FWCI
- 25.97
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- neural information processing systems (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- National Natural Science Foundation of China (62176152)
- National Key Research and Development Program of China (2021YFB1716000)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Anomaly Detection Techniques and Applications
- Physical Sciences › Computer Science › Artificial Intelligence › Imbalanced Data Classification Techniques
- Physical Sciences › Computer Science › Artificial Intelligence › Adversarial Robustness in Machine Learning
- Keywords
- multi-class anomaly detection, unified model, anomaly detection
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-29, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-29 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (qwen3.8-omni-flash and gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.