# Cross-domain prompt generalization on ImageNet variants (test)

Source: [Cross-domain prompt generalization on ImageNet variants (test)](<https://www.sota2.com/research/sota/cross-domain-prompt-generalization-on-imagenet-variants-test>)

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

Data updated: 2026-02-26T05:26:51.207Z

Evaluation of model performance across multiple ImageNet variants (V2, Sketch, A, R) using cross-domain prompts, reporting performance metrics for each variant and an overall average.

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 | Cross-domain prompt generalization |
| Dataset (including any stated split) | ImageNet to ImageNet variants (test) |
| Metric | Acc (ImageNetV2) |
| 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

- [Acc (ImageNetV2)](<https://www.sota2.com/research/sota/cross-domain-prompt-generalization-on-imagenet-variants-test.md?metric=Acc%20(ImageNetV2)>): Higher is better; unit: PERCENT.

- [Acc (ImageNet-Sketch)](<https://www.sota2.com/research/sota/cross-domain-prompt-generalization-on-imagenet-variants-test.md?metric=Acc%20(ImageNet-Sketch)>): Higher is better; unit: PERCENT.

- [Acc (ImageNet-A)](<https://www.sota2.com/research/sota/cross-domain-prompt-generalization-on-imagenet-variants-test.md?metric=Acc%20(ImageNet-A)>): Higher is better; unit: PERCENT.

- [Acc (ImageNet-R)](<https://www.sota2.com/research/sota/cross-domain-prompt-generalization-on-imagenet-variants-test.md?metric=Acc%20(ImageNet-R)>): Higher is better; unit: PERCENT.

- [Avg Acc](<https://www.sota2.com/research/sota/cross-domain-prompt-generalization-on-imagenet-variants-test.md?metric=Avg%20Acc>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Acc (ImageNetV2) | Result date | Report paper |
| --- | --- | --- | --- |
| Bayesian Prompt Learning | 64.23 | Not recorded | 68fa70571e209131cea4a691 |
| CoOp | 64.2 | Not recorded | 68fa70571e209131cea4a691 |
| CoCoOp | 64.07 | Not recorded | 68fa70571e209131cea4a691 |

### Bayesian Prompt Learning

Acc (ImageNetV2): 64.23. Result ID: 699114e73a6c314849058988.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70571e209131cea4a691.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70571e209131cea4a691.md>)

[Implementation](<https://github.com/saic-fi/Bayesian-Prompt-Learning>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### CoOp

Acc (ImageNetV2): 64.2. Result ID: 699114e73a6c314849058986.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70571e209131cea4a691.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d8b1e209131cea3a814.md>)

[Implementation](<https://github.com/KaiyangZhou/CoOp>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### CoCoOp

Acc (ImageNetV2): 64.07. Result ID: 699114e73a6c314849058987.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70571e209131cea4a691.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d8c1e209131cea3a831.md>)

[Implementation](<https://github.com/KaiyangZhou/CoOp>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

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