# Image Classification on 9 Vision Tasks + ImageNet-1K (test)

Source: [Image Classification on 9 Vision Tasks + ImageNet-1K (test)](<https://www.sota2.com/research/sota/image-classification-on-9-vision-tasks-plus-imagenet-1k-test>)

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

Data updated: 2026-02-26T06:03:33.822Z

Evaluation of Image Classification performance across a suite of 9 diverse vision tasks (including SUN397, Cars, MNIST, etc.) plus ImageNet-1K, reported on the test split, with results aggregated via average accuracy.

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 | Image Classification |
| Dataset (including any stated split) | 9 Vision Tasks ImageNet-1K added (test) |
| Metric | SUN397 Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| 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

- [SUN397 Accuracy](<https://www.sota2.com/research/sota/image-classification-on-9-vision-tasks-plus-imagenet-1k-test.md?metric=SUN397%20Accuracy>): Higher is better; unit: PERCENT.

- [Cars Accuracy](<https://www.sota2.com/research/sota/image-classification-on-9-vision-tasks-plus-imagenet-1k-test.md?metric=Cars%20Accuracy>): Higher is better; unit: PERCENT.

- [RESISC45 Accuracy](<https://www.sota2.com/research/sota/image-classification-on-9-vision-tasks-plus-imagenet-1k-test.md?metric=RESISC45%20Accuracy>): Higher is better; unit: PERCENT.

- [EuroSAT Accuracy](<https://www.sota2.com/research/sota/image-classification-on-9-vision-tasks-plus-imagenet-1k-test.md?metric=EuroSAT%20Accuracy>): Higher is better; unit: PERCENT.

- [SVHN Accuracy](<https://www.sota2.com/research/sota/image-classification-on-9-vision-tasks-plus-imagenet-1k-test.md?metric=SVHN%20Accuracy>): Higher is better; unit: PERCENT.

- [GTSRB Accuracy](<https://www.sota2.com/research/sota/image-classification-on-9-vision-tasks-plus-imagenet-1k-test.md?metric=GTSRB%20Accuracy>): Higher is better; unit: PERCENT.

- [MNIST Accuracy](<https://www.sota2.com/research/sota/image-classification-on-9-vision-tasks-plus-imagenet-1k-test.md?metric=MNIST%20Accuracy>): Higher is better; unit: PERCENT.

- [DTD Accuracy](<https://www.sota2.com/research/sota/image-classification-on-9-vision-tasks-plus-imagenet-1k-test.md?metric=DTD%20Accuracy>): Higher is better; unit: PERCENT.

- [ImageNet-1K Accuracy](<https://www.sota2.com/research/sota/image-classification-on-9-vision-tasks-plus-imagenet-1k-test.md?metric=ImageNet-1K%20Accuracy>): Higher is better; unit: PERCENT.

- [Average Accuracy](<https://www.sota2.com/research/sota/image-classification-on-9-vision-tasks-plus-imagenet-1k-test.md?metric=Average%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | SUN397 Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| EMR-MERGING | 77 | Not recorded | 68fa74f31e209131cea64c99 |
| Individual | 75.3 | Not recorded | 68fa74f31e209131cea64c99 |
| Weight Averaging | 61.8 | Not recorded | 68fa74f31e209131cea64c99 |
| Ties-Merging | 53.3 | Not recorded | 68fa74f31e209131cea64c99 |
| Task Arithmetic | 51.8 | Not recorded | 68fa74f31e209131cea64c99 |

### EMR-MERGING

SUN397 Accuracy: 77. Result ID: 69936be43a6c31484915ce33.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74f31e209131cea64c99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74f31e209131cea64c99.md>)

[Implementation](<https://github.com/harveyhuang18/EMR_Merging>)

Recorded evaluation configuration: {"Backbone":"ViT-B/32"}. This record does not assert independent reproduction.

### Individual

SUN397 Accuracy: 75.3. Result ID: 69936be43a6c31484915ce2f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74f31e209131cea64c99.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B/32"}. This record does not assert independent reproduction.

### Weight Averaging

SUN397 Accuracy: 61.8. Result ID: 69936be43a6c31484915ce30.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74f31e209131cea64c99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ce11e209131cea3679c.md>)

[Implementation](<https://github.com/mlfoundations/model-soups>)

Recorded evaluation configuration: {"Backbone":"ViT-B/32"}. This record does not assert independent reproduction.

### Ties-Merging

SUN397 Accuracy: 53.3. Result ID: 69936be43a6c31484915ce32.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74f31e209131cea64c99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71091e209131cea4ecd8.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B/32"}. This record does not assert independent reproduction.

### Task Arithmetic

SUN397 Accuracy: 51.8. Result ID: 69936be43a6c31484915ce31.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74f31e209131cea64c99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ef01e209131cea43012.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B/32"}. This record does not assert independent reproduction.

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