# Zero-shot classification on FACET

Source: [Zero-shot classification on FACET](<https://www.sota2.com/research/sota/zero-shot-classification-on-facet>)

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

Data updated: 2026-02-26T05:51:51.959Z

Evaluation of Zero-shot Classification performance on the FACET dataset, using metrics like Mean Accuracy (Mean Acc.), Mean DP, and Max DP.

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 | Zero-shot classification |
| Dataset (including any stated split) | FACET |
| Metric | Mean Accuracy |
| 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

- [Mean Accuracy](<https://www.sota2.com/research/sota/zero-shot-classification-on-facet.md?metric=Mean%20Accuracy>): Higher is better; unit: PERCENT.

- [Mean DP](<https://www.sota2.com/research/sota/zero-shot-classification-on-facet.md?metric=Mean%20DP>): Higher is better; unit: PERCENT.

- [Max DP](<https://www.sota2.com/research/sota/zero-shot-classification-on-facet.md?metric=Max%20DP>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Mean Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| XVLM | 56.61 | Not recorded | 68fa756b1e209131cea67c07 |
| XVLM | 56.51 | Not recorded | 68fa756b1e209131cea67c07 |
| CLIP-ViT-B/32 | 52.48 | Not recorded | 68fa756b1e209131cea67c07 |
| CLIP-ViT-B/32 | 51.97 | Not recorded | 68fa756b1e209131cea67c07 |
| CLIP-RN50 | 51.92 | Not recorded | 68fa756b1e209131cea67c07 |
| CLIP-RN50 | 51.35 | Not recorded | 68fa756b1e209131cea67c07 |

### XVLM

Mean Accuracy: 56.61. Result ID: 6993acb23a6c31484917bb85.

[Reporting paper](<https://www.sota2.com/research/paper/68fa756b1e209131cea67c07.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa756b1e209131cea67c07.md>)

Recorded evaluation configuration: {"Debiasing Variant":"Baseline"}. This record does not assert independent reproduction.

### XVLM

Mean Accuracy: 56.51. Result ID: 6993acb23a6c31484917bb86.

[Reporting paper](<https://www.sota2.com/research/paper/68fa756b1e209131cea67c07.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa756b1e209131cea67c07.md>)

Recorded evaluation configuration: {"Debiasing Variant":"SFID"}. This record does not assert independent reproduction.

### CLIP-ViT-B/32

Mean Accuracy: 52.48. Result ID: 6993acb23a6c31484917bb83.

[Reporting paper](<https://www.sota2.com/research/paper/68fa756b1e209131cea67c07.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa756b1e209131cea67c07.md>)

Recorded evaluation configuration: {"Debiasing Variant":"Baseline"}. This record does not assert independent reproduction.

### CLIP-ViT-B/32

Mean Accuracy: 51.97. Result ID: 6993acb23a6c31484917bb84.

[Reporting paper](<https://www.sota2.com/research/paper/68fa756b1e209131cea67c07.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa756b1e209131cea67c07.md>)

Recorded evaluation configuration: {"Debiasing Variant":"SFID"}. This record does not assert independent reproduction.

### CLIP-RN50

Mean Accuracy: 51.92. Result ID: 6993acb23a6c31484917bb81.

[Reporting paper](<https://www.sota2.com/research/paper/68fa756b1e209131cea67c07.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa756b1e209131cea67c07.md>)

Recorded evaluation configuration: {"Debiasing Variant":"Baseline"}. This record does not assert independent reproduction.

### CLIP-RN50

Mean Accuracy: 51.35. Result ID: 6993acb23a6c31484917bb82.

[Reporting paper](<https://www.sota2.com/research/paper/68fa756b1e209131cea67c07.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa756b1e209131cea67c07.md>)

Recorded evaluation configuration: {"Debiasing Variant":"SFID"}. This record does not assert independent reproduction.

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