# Multi-label Classification on MS-COCO 2014

Source: [Multi-label Classification on MS-COCO 2014](<https://www.sota2.com/research/sota/multi-label-classification-on-ms-coco-2014>)

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

Data updated: 2026-02-27T11:46:09.692Z

Evaluation of Multi-label Classification performance on the MS-COCO 2014 dataset, reporting Precision, Recall, and F1 scores at top-3 and top-5 predictions, alongside mAP.

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 | Multi-label Classification |
| Dataset (including any stated split) | MS-COCO 2014 |
| Metric | Precision @ 3 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 6 of 6 |
| 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

- [Precision @ 3](<https://www.sota2.com/research/sota/multi-label-classification-on-ms-coco-2014.md?metric=Precision%20%40%203>): Higher is better; unit: PERCENT.

- [Recall @ 3](<https://www.sota2.com/research/sota/multi-label-classification-on-ms-coco-2014.md?metric=Recall%20%40%203>): Higher is better; unit: PERCENT.

- [F1 Score @ 3](<https://www.sota2.com/research/sota/multi-label-classification-on-ms-coco-2014.md?metric=F1%20Score%20%40%203>): Higher is better; unit: PERCENT.

- [Precision @ 5](<https://www.sota2.com/research/sota/multi-label-classification-on-ms-coco-2014.md?metric=Precision%20%40%205>): Higher is better; unit: PERCENT.

- [Recall @ 5](<https://www.sota2.com/research/sota/multi-label-classification-on-ms-coco-2014.md?metric=Recall%20%40%205>): Higher is better; unit: PERCENT.

- [F1 Score @ 5](<https://www.sota2.com/research/sota/multi-label-classification-on-ms-coco-2014.md?metric=F1%20Score%20%40%205>): Higher is better; unit: PERCENT.

- [Mean Average Precision (mAP)](<https://www.sota2.com/research/sota/multi-label-classification-on-ms-coco-2014.md?metric=Mean%20Average%20Precision%20(mAP)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Precision @ 3 | Result date | Report paper |
| --- | --- | --- | --- |
| UCAT | 36.58 | Not recorded | 6940e7a8456146936284f897 |
| FARE | 33.45 | Not recorded | 6940e7a8456146936284f897 |
| TGA-ZSR | 32.95 | Not recorded | 6940e7a8456146936284f897 |
| PMG-AFT | 32.32 | Not recorded | 6940e7a8456146936284f897 |
| TeCoA | 30.23 | Not recorded | 6940e7a8456146936284f897 |
| CLIP | 17.72 | Not recorded | 6940e7a8456146936284f897 |

### UCAT

Precision @ 3: 36.58. Result ID: 69a183f135dfcabb00e78847.

[Reporting paper](<https://www.sota2.com/research/paper/6940e7a8456146936284f897.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6940e7a8456146936284f897.md>)

Recorded evaluation configuration: {"Backbone":"CLIP-B/32","Evaluation Mode":"Zero-shot","Evaluation Attack":"CW-100","Training Attack":"10-step PGD (ε = 2/255)","Training Dataset":"TinyImageNet"}. This record does not assert independent reproduction.

### FARE

Precision @ 3: 33.45. Result ID: 69a183f135dfcabb00e78844.

[Reporting paper](<https://www.sota2.com/research/paper/6940e7a8456146936284f897.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa738c1e209131cea5d24a.md>)

Recorded evaluation configuration: {"Backbone":"CLIP-B/32","Evaluation Mode":"Zero-shot","Evaluation Attack":"CW-100","Training Attack":"10-step PGD (ε = 2/255)","Training Dataset":"TinyImageNet"}. This record does not assert independent reproduction.

### TGA-ZSR

Precision @ 3: 32.95. Result ID: 69a183f135dfcabb00e78846.

[Reporting paper](<https://www.sota2.com/research/paper/6940e7a8456146936284f897.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75771e209131cea68256.md>)

[Implementation](<https://github.com/zhyblue424/TGA-ZSR>)

Recorded evaluation configuration: {"Backbone":"CLIP-B/32","Evaluation Mode":"Zero-shot","Evaluation Attack":"CW-100","Training Attack":"10-step PGD (ε = 2/255)","Training Dataset":"TinyImageNet"}. This record does not assert independent reproduction.

### PMG-AFT

Precision @ 3: 32.32. Result ID: 69a183f135dfcabb00e78845.

[Reporting paper](<https://www.sota2.com/research/paper/6940e7a8456146936284f897.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72df1e209131cea5912e.md>)

Recorded evaluation configuration: {"Backbone":"CLIP-B/32","Evaluation Mode":"Zero-shot","Evaluation Attack":"CW-100","Training Attack":"10-step PGD (ε = 2/255)","Training Dataset":"TinyImageNet"}. This record does not assert independent reproduction.

### TeCoA

Precision @ 3: 30.23. Result ID: 69a183f135dfcabb00e78843.

[Reporting paper](<https://www.sota2.com/research/paper/6940e7a8456146936284f897.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f271e209131cea44177.md>)

Recorded evaluation configuration: {"Backbone":"CLIP-B/32","Evaluation Mode":"Zero-shot","Evaluation Attack":"CW-100","Training Attack":"10-step PGD (ε = 2/255)","Training Dataset":"TinyImageNet"}. This record does not assert independent reproduction.

### CLIP

Precision @ 3: 17.72. Result ID: 69a183f135dfcabb00e78842.

[Reporting paper](<https://www.sota2.com/research/paper/6940e7a8456146936284f897.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69581e209131cea21ad0.md>)

[Implementation](<https://github.com/OpenAI/CLIP>)

Recorded evaluation configuration: {"Backbone":"CLIP-B/32","Evaluation Mode":"Zero-shot","Evaluation Attack":"CW-100","Training Attack":"10-step PGD (ε = 2/255)","Training Dataset":"TinyImageNet"}. This record does not assert independent reproduction.

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