Papers › Dinomaly: The Less Is More Philosophy in Multi-Class Unsupervised Anomaly Detection

Dinomaly: The Less Is More Philosophy in Multi-Class Unsupervised Anomaly Detection

23 May 2024CVPR 2025 1arXiv:2405.14325archive 2025-07-28

Jia Guo, Shuai Lu, Weihang Zhang, Fang Chen, Hongen Liao, Huiqi Li

Recent studies highlighted a practical setting of unsupervised anomaly detection (UAD) that builds a unified model for multi-class images. Despite various advancements addressing this challenging task, the detection performance under the multi-class setting still lags far behind state-of-the-art class-separated models. Our research aims to bridge this substantial performance gap. In this paper, we introduce a minimalistic reconstruction-based anomaly detection framework, namely Dinomaly, which leverages pure Transformer architectures without relying on complex designs, additional modules, or specialized tricks. Given this powerful framework consisted of only Attentions and MLPs, we found four simple components that are essential to multi-class anomaly detection: (1) Foundation Transformers that extracts universal and discriminative features, (2) Noisy Bottleneck where pre-existing Dropouts do all the noise injection tricks, (3) Linear Attention that naturally cannot focus, and (4) Loose Reconstruction that does not force layer-to-layer and point-by-point reconstruction. Extensive experiments are conducted across popular anomaly detection benchmarks including MVTec-AD, VisA, and Real-IAD. Our proposed Dinomaly achieves impressive image-level AUROC of 99.6%, 98.7%, and 89.3% on the three datasets respectively, which is not only superior to state-of-the-art multi-class UAD methods, but also achieves the most advanced class-separated UAD records.

PaperPDFConference PDFCodeCode Syntology ran

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2405.14325")

Code

Syntology Ran 9 of 10 code samples harvested from 1 repository linked to this paper; 1 has no recorded run. Of those that ran: 1 ran · honoured contract; 4 ran · our draft was wrong; 1 ran · fixture could not drive it; 3 ran with no contract checked.

By repository: official repository: 10 samples from 1 repository, 9 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

guojiajeremy/dinomaly officialmentioned in papermentioned on GitHubpytorchApache-2.0 report
septmars/DL pytorchMIT report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

10 samples harvested; 9 ran; 1 honoured the contract we drafted; 1 has no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

1ran · honoured contract
4ran · our draft was wrong
1ran · fixture could not drive it
3ran
1unverified

Licence: 0 of the 10 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from guojiajeremy/dinomaly. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

conv1x1 guojiajeremy/dinomaly/models/de_resnet.py official repository ran · our draft was wrong Apache-2.0 (permissive) · 2a80220dabcb742a · report
conv3x3 guojiajeremy/dinomaly/models/de_resnet.py official repository ran · our draft was wrong Apache-2.0 (permissive) · 600ff2c45e0de056 · report
count_parameters guojiajeremy/dinomaly/dinomaly_mpdd_sep.py official repository ran · honoured contract Apache-2.0 (permissive) · f6b944f50d3f15ae · report
deconv2x2 guojiajeremy/dinomaly/models/de_resnet.py official repository ran · our draft was wrong Apache-2.0 (permissive) · 8d035278d5e4a82a · report
drop_path guojiajeremy/dinomaly/models/vision_transformer.py official repository ran · fixture could not drive it fingerprinted Apache-2.0 (permissive) · 55120f2026b56aa2 · report
get_data_transforms guojiajeremy/dinomaly/dataset.py official repository ran Apache-2.0 (permissive) · 98f14d98f255b3b2 · report
get_logger guojiajeremy/dinomaly/dinomaly_mpdd_sep.py official repository ran · our draft was wrong Apache-2.0 (permissive) · a1db6c55cfde6e19 · report
get_strong_transforms guojiajeremy/dinomaly/dataset.py official repository ran Apache-2.0 (permissive) · 8a9c4b925a7212c9 · report
update_average guojiajeremy/dinomaly/models/uad.py official repository ran fingerprinted Apache-2.0 (permissive) · dd357f5ea5a0a8a5 · report
cdconv3x3 guojiajeremy/dinomaly/models/resnet.py official repository unverified Apache-2.0 (permissive) · ef97f3062faf843d · report

Tasks

Anomaly DetectionMulti-class Anomaly DetectionPhilosophyUnsupervised Anomaly Detection

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Anomaly Detection MPDD Dinomaly Detection AUROC 97.2 #7 of 16 Archive leaderboard report
Anomaly Detection MPDD Dinomaly Segmentation AUROC 99.1 #7 of 16 Archive leaderboard report
Anomaly Detection MVTec AD Dinomaly ViT-L (model-unified multi-class) Detection AUROC 99.77 #8 of 148 Archive leaderboard report
Anomaly Detection MVTec AD Dinomaly ViT-L (model-unified multi-class) Segmentation AP 70.53 #8 of 148 Archive leaderboard report
Anomaly Detection MVTec AD Dinomaly ViT-L (model-unified multi-class) Segmentation AUPRO 95.09 #8 of 148 Archive leaderboard report
Anomaly Detection MVTec AD Dinomaly ViT-L (model-unified multi-class) Segmentation AUROC 98.54 #8 of 148 Archive leaderboard report
Anomaly Detection MVTec AD Dinomaly ViT-B (model-unified multi-class) Detection AUROC 99.60 #19 of 148 Archive leaderboard report
Anomaly Detection MVTec AD Dinomaly ViT-B (model-unified multi-class) Segmentation AP 69.29 #19 of 148 Archive leaderboard report
Anomaly Detection MVTec AD Dinomaly ViT-B (model-unified multi-class) Segmentation AUPRO 94.79 #19 of 148 Archive leaderboard report
Anomaly Detection MVTec AD Dinomaly ViT-B (model-unified multi-class) Segmentation AUROC 98.35 #19 of 148 Archive leaderboard report
Anomaly Detection VisA Dinomaly ViT-L (model-unified multi-class) Detection AUROC 98.9 #4 of 50 Archive leaderboard report
Anomaly Detection VisA Dinomaly ViT-L (model-unified multi-class) F1-Score 96.1 #4 of 50 Archive leaderboard report
Anomaly Detection VisA Dinomaly ViT-L (model-unified multi-class) Segmentation AUPRO 94.8 #4 of 50 Archive leaderboard report
Anomaly Detection VisA Dinomaly ViT-L (model-unified multi-class) Segmentation AUPRO (until 30% FPR) 94.8 #4 of 50 Archive leaderboard report
Anomaly Detection VisA Dinomaly ViT-L (model-unified multi-class) Segmentation AUROC 99.1 #4 of 50 Archive leaderboard report
Multi-class Anomaly Detection MVTec AD Dinomaly-Large Detection AUROC 99.8 #2 of 13 Archive leaderboard report
Multi-class Anomaly Detection MVTec AD Dinomaly-Large Segmentation AUROC 98.5 #2 of 13 Archive leaderboard report
Multi-class Anomaly Detection MVTec AD Dinomaly-Base Detection AUROC 99.6 #4 of 13 Archive leaderboard report
Multi-class Anomaly Detection MVTec AD Dinomaly-Base Segmentation AUROC 98.4 #4 of 13 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Absolute Position EncodingsAdamAttentionBPEDense ConnectionsDropoutLabel SmoothingLayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSoftmaxTransformer

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections