# Image Classification on ImageNet400 (200+200) 1k-subset (test)

Source: [Image Classification on ImageNet400 (200+200) 1k-subset (test)](<https://www.sota2.com/research/sota/image-classification-on-imagenet400-200-plus-200-1k-subset-test>)

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

Data updated: 2026-02-26T05:47:39.570Z

Evaluation of Image Classification performance on the specific ImageNet400 (200+200) 1k-subset test split, reporting accuracy metrics.

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) | ImageNet400 (200+200) 1k-subset (test) |
| Metric | Joint Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 7 of 7 |
| 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

- [Joint Accuracy](<https://www.sota2.com/research/sota/image-classification-on-imagenet400-200-plus-200-1k-subset-test.md?metric=Joint%20Accuracy>): Higher is better; unit: PERCENT.

- [Avg Accuracy](<https://www.sota2.com/research/sota/image-classification-on-imagenet400-200-plus-200-1k-subset-test.md?metric=Avg%20Accuracy>): Higher is better; unit: PERCENT.

- [T. A Accuracy](<https://www.sota2.com/research/sota/image-classification-on-imagenet400-200-plus-200-1k-subset-test.md?metric=T.%20A%20Accuracy>): Higher is better; unit: PERCENT.

- [T. B Accuracy](<https://www.sota2.com/research/sota/image-classification-on-imagenet400-200-plus-200-1k-subset-test.md?metric=T.%20B%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Joint Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| MuDSC\_Align | 44.87 | Not recorded | 68fa727f1e209131cea56c4e |
| A. Align | 44.86 | Not recorded | 68fa727f1e209131cea56c4e |
| MuDSC\_Zip | 44.86 | Not recorded | 68fa727f1e209131cea56c4e |
| Average | 44.85 | Not recorded | 68fa727f1e209131cea56c4e |
| Rebasin | 44.85 | Not recorded | 68fa727f1e209131cea56c4e |
| W.Zip | 44.85 | Not recorded | 68fa727f1e209131cea56c4e |
| Zipit | 44.22 | Not recorded | 68fa727f1e209131cea56c4e |

### MuDSC\_Align

Joint Accuracy: 44.87. Result ID: 69922e503a6c3148490d1f4c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa727f1e209131cea56c4e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa727f1e209131cea56c4e.md>)

[Implementation](<https://github.com/zju-vipa/training_free_model_merging>)

Recorded evaluation configuration: {"Backbone":"ResNet-50","Pre-training":"DINO"}. This record does not assert independent reproduction.

### A. Align

Joint Accuracy: 44.86. Result ID: 69922e503a6c3148490d1f4b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa727f1e209131cea56c4e.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-50","Pre-training":"DINO"}. This record does not assert independent reproduction.

### MuDSC\_Zip

Joint Accuracy: 44.86. Result ID: 69922e503a6c3148490d1f4f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa727f1e209131cea56c4e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa727f1e209131cea56c4e.md>)

[Implementation](<https://github.com/zju-vipa/training_free_model_merging>)

Recorded evaluation configuration: {"Backbone":"ResNet-50","Pre-training":"DINO"}. This record does not assert independent reproduction.

### Average

Joint Accuracy: 44.85. Result ID: 69922e503a6c3148490d1f49.

[Reporting paper](<https://www.sota2.com/research/paper/68fa727f1e209131cea56c4e.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-50","Pre-training":"DINO"}. This record does not assert independent reproduction.

### Rebasin

Joint Accuracy: 44.85. Result ID: 69922e503a6c3148490d1f4a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa727f1e209131cea56c4e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ea91e209131cea412f7.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-50","Pre-training":"DINO"}. This record does not assert independent reproduction.

### W.Zip

Joint Accuracy: 44.85. Result ID: 69922e503a6c3148490d1f4e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa727f1e209131cea56c4e.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-50","Pre-training":"DINO"}. This record does not assert independent reproduction.

### Zipit

Joint Accuracy: 44.22. Result ID: 69922e503a6c3148490d1f4d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa727f1e209131cea56c4e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72811e209131cea56dce.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-50","Pre-training":"DINO"}. This record does not assert independent reproduction.

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