# Open vocabulary image classification on Wiki-H

Source: [Open vocabulary image classification on Wiki-H](<https://www.sota2.com/research/sota/open-vocabulary-image-classification-on-wiki-h>)

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

Data updated: 2026-02-26T06:07:55.847Z

Evaluation of Open Vocabulary Image Classification performance on the Wiki-H dataset, measured using metrics like Primary Score, Prediction Score, and Overall Score.

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 | Open vocabulary image classification |
| Dataset (including any stated split) | Wiki-H |
| Metric | Used Objects |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 10 of 10 |
| 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

- [Used Objects](<https://www.sota2.com/research/sota/open-vocabulary-image-classification-on-wiki-h.md?metric=Used%20Objects>): Higher is better; unit: SCALAR.

- [RAM Vocab Size](<https://www.sota2.com/research/sota/open-vocabulary-image-classification-on-wiki-h.md?metric=RAM%20Vocab%20Size>): Higher is better; unit: SCALAR.

- [Primary Score](<https://www.sota2.com/research/sota/open-vocabulary-image-classification-on-wiki-h.md?metric=Primary%20Score>): Higher is better; unit: PERCENT.

- [Prediction Score](<https://www.sota2.com/research/sota/open-vocabulary-image-classification-on-wiki-h.md?metric=Prediction%20Score>): Higher is better; unit: PERCENT.

- [Overall Score](<https://www.sota2.com/research/sota/open-vocabulary-image-classification-on-wiki-h.md?metric=Overall%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Used Objects | Result date | Report paper |
| --- | --- | --- | --- |
| NOVIC | 934 | Not recorded | 68fa77fe1e209131cea7647c |
| NOVIC | 893 | Not recorded | 68fa77fe1e209131cea7647c |
| NOVIC | 852 | Not recorded | 68fa77fe1e209131cea7647c |
| NOVIC | 846 | Not recorded | 68fa77fe1e209131cea7647c |
| NOVIC | 794 | Not recorded | 68fa77fe1e209131cea7647c |
| NOVIC | 693 | Not recorded | 68fa77fe1e209131cea7647c |
| RAM | 399 | Not recorded | 68fa77fe1e209131cea7647c |
| Tag2Text | 397 | Not recorded | 68fa77fe1e209131cea7647c |
| RAM | 391 | Not recorded | 68fa77fe1e209131cea7647c |
| RAM | 348 | Not recorded | 68fa77fe1e209131cea7647c |

### NOVIC

Used Objects: 934. Result ID: 699577b53badfaffe66e3857.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77fe1e209131cea7647c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77fe1e209131cea7647c.md>)

Recorded evaluation configuration: {"Seen Objects":"42919","Backbone":"DFN-5B H/14-378 CLIP","Frequency Threshold":"FT0","Evaluation":"Human annotations"}. This record does not assert independent reproduction.

### NOVIC

Used Objects: 893. Result ID: 699577b53badfaffe66e3856.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77fe1e209131cea7647c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77fe1e209131cea7647c.md>)

Recorded evaluation configuration: {"Seen Objects":"11897","Backbone":"DFN-5B H/14-378 CLIP","Frequency Threshold":"FT2","Evaluation":"Human annotations"}. This record does not assert independent reproduction.

### NOVIC

Used Objects: 852. Result ID: 699577b53badfaffe66e3855.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77fe1e209131cea7647c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77fe1e209131cea7647c.md>)

Recorded evaluation configuration: {"Seen Objects":"42919","Backbone":"DFN-5B H/14-378 CLIP","Frequency Threshold":"FT0"}. This record does not assert independent reproduction.

### NOVIC

Used Objects: 846. Result ID: 699577b53badfaffe66e3854.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77fe1e209131cea7647c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77fe1e209131cea7647c.md>)

Recorded evaluation configuration: {"Seen Objects":"11897","Backbone":"DFN-5B H/14-378 CLIP","Frequency Threshold":"FT2"}. This record does not assert independent reproduction.

### NOVIC

Used Objects: 794. Result ID: 699577b53badfaffe66e3853.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77fe1e209131cea7647c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77fe1e209131cea7647c.md>)

Recorded evaluation configuration: {"Seen Objects":"5899","Backbone":"DFN-5B H/14-378 CLIP","Frequency Threshold":"FT6"}. This record does not assert independent reproduction.

### NOVIC

Used Objects: 693. Result ID: 699577b53badfaffe66e3852.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77fe1e209131cea7647c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77fe1e209131cea7647c.md>)

Recorded evaluation configuration: {"Seen Objects":"2919","Backbone":"DFN-5B H/14-378 CLIP","Frequency Threshold":"FT9"}. This record does not assert independent reproduction.

### RAM

Used Objects: 399. Result ID: 699577b53badfaffe66e3851.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77fe1e209131cea7647c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fbc1e209131cea471ee.md>)

Recorded evaluation configuration: {"Seen Objects":"42919","Backbone":"DFN-5B H/14-378 CLIP","Frequency Threshold":"FT0"}. This record does not assert independent reproduction.

### Tag2Text

Used Objects: 397. Result ID: 699577b43badfaffe66e384e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77fe1e209131cea7647c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72921e209131cea572fa.md>)

Recorded evaluation configuration: {"Seen Objects":"3429","Backbone":"DFN-5B H/14-378 CLIP"}. This record does not assert independent reproduction.

### RAM

Used Objects: 391. Result ID: 699577b53badfaffe66e3850.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77fe1e209131cea7647c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fbc1e209131cea471ee.md>)

Recorded evaluation configuration: {"Seen Objects":"11897","Backbone":"DFN-5B H/14-378 CLIP","Frequency Threshold":"FT2"}. This record does not assert independent reproduction.

### RAM

Used Objects: 348. Result ID: 699577b53badfaffe66e384f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77fe1e209131cea7647c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fbc1e209131cea471ee.md>)

Recorded evaluation configuration: {"Seen Objects":"4585","Backbone":"DFN-5B H/14-378 CLIP"}. This record does not assert independent reproduction.

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