# Anomaly Localization on Avg (MVTec-AD, MPDD, VisA, PCB-Bank) (test)

Source: [Anomaly Localization on Avg (MVTec-AD, MPDD, VisA, PCB-Bank) (test)](<https://www.sota2.com/research/sota/anomaly-localization-on-avg-mvtec-ad-mpdd-visa-pcb-bank-test>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T05:54:40.387Z

Evaluation of anomaly localization performance averaged across the test splits of MVTec-AD, MPDD, VisA, and PCB-Bank datasets, using metrics like P-AUROC, P-AP, P-F1-max, and PRO.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Anomaly Localization |
| Dataset (including any stated split) | Avg (MVTec-AD, MPDD, VisA, PCB-Bank) (test) |
| Metric | P-AUROC |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 3 of 3 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [P-AUROC](<https://www.sota2.com/research/sota/anomaly-localization-on-avg-mvtec-ad-mpdd-visa-pcb-bank-test.md?metric=P-AUROC>): Higher is better; unit: PERCENT.

- [P-AP](<https://www.sota2.com/research/sota/anomaly-localization-on-avg-mvtec-ad-mpdd-visa-pcb-bank-test.md?metric=P-AP>): Higher is better; unit: PERCENT.

- [P-F1-max](<https://www.sota2.com/research/sota/anomaly-localization-on-avg-mvtec-ad-mpdd-visa-pcb-bank-test.md?metric=P-F1-max>): Higher is better; unit: PERCENT.

- [PRO](<https://www.sota2.com/research/sota/anomaly-localization-on-avg-mvtec-ad-mpdd-visa-pcb-bank-test.md?metric=PRO>): Higher is better; unit: PERCENT.

## Recorded results

| Method | P-AUROC | Result date | Report paper |
| --- | --- | --- | --- |
| GLAD | 98 | Not recorded | 68fa74bb1e209131cea63561 |
| UniAD | 96.3 | Not recorded | 68fa74bb1e209131cea63561 |
| DiAD | 96.1 | Not recorded | 68fa74bb1e209131cea63561 |

### GLAD

P-AUROC: 98. Result ID: 699355d03a6c314849153fa5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74bb1e209131cea63561.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74bb1e209131cea63561.md>)

[Implementation](<https://github.com/hyao1/GLAD>)

Recorded evaluation configuration: {"resolution":"256","setting":"multi-category"}. This record does not assert independent reproduction.

### UniAD

P-AUROC: 96.3. Result ID: 699355d03a6c314849153fa3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74bb1e209131cea63561.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6dba1e209131cea3b9bd.md>)

[Implementation](<https://github.com/zhiyuanyou/UniAD>)

Recorded evaluation configuration: {"setting":"multi-category"}. This record does not assert independent reproduction.

### DiAD

P-AUROC: 96.1. Result ID: 699355d03a6c314849153fa4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74bb1e209131cea63561.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77e31e209131cea75d18.md>)

Recorded evaluation configuration: {"setting":"multi-category"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.