# Image Classification on ImageNet distribution shift 1k A K R V (val test)

Source: [Image Classification on ImageNet distribution shift 1k A K R V (val test)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-distribution-shift-1k-a-k-r-v-val-test>)

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

Data updated: 2026-02-26T06:06:53.607Z

Evaluation of Image Classification performance across the specific set of ImageNet distribution shift variants: ImageNet (reference), ImageNet-A, ImageNet-K, ImageNet-R, and ImageNet-V, using the combined validation and test splits, reporting individual and aggregate scores.

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) | ImageNet distribution shift 1k A K R V (val test) |
| Metric | Top-1 Acc (ImageNet) |
| Unit | PERCENT |
| 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

- [Top-1 Acc (ImageNet)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-distribution-shift-1k-a-k-r-v-val-test.md?metric=Top-1%20Acc%20(ImageNet)>): Higher is better; unit: PERCENT.

- [Top-1 Acc (ImageNet-A)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-distribution-shift-1k-a-k-r-v-val-test.md?metric=Top-1%20Acc%20(ImageNet-A)>): Higher is better; unit: PERCENT.

- [Top-1 Acc (ImageNet-K)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-distribution-shift-1k-a-k-r-v-val-test.md?metric=Top-1%20Acc%20(ImageNet-K)>): Higher is better; unit: PERCENT.

- [Top-1 Acc (ImageNet-R)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-distribution-shift-1k-a-k-r-v-val-test.md?metric=Top-1%20Acc%20(ImageNet-R)>): Higher is better; unit: PERCENT.

- [Top-1 Acc (ImageNet-V)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-distribution-shift-1k-a-k-r-v-val-test.md?metric=Top-1%20Acc%20(ImageNet-V)>): Higher is better; unit: PERCENT.

- [Avg. Top-1 Acc](<https://www.sota2.com/research/sota/image-classification-on-imagenet-distribution-shift-1k-a-k-r-v-val-test.md?metric=Avg.%20Top-1%20Acc>): Higher is better; unit: PERCENT.

- [Avg. Top-1 Acc (OOD)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-distribution-shift-1k-a-k-r-v-val-test.md?metric=Avg.%20Top-1%20Acc%20(OOD)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Top-1 Acc (ImageNet) | Result date | Report paper |
| --- | --- | --- | --- |
| FCL (Ours) | 65.55 | Not recorded | 698abba9c97a0b3e3b5ddff6 |
| ZERO | 65.29 | Not recorded | 698abba9c97a0b3e3b5ddff6 |
| MTA | 65.02 | Not recorded | 698abba9c97a0b3e3b5ddff6 |
| TTL | 64.99 | Not recorded | 698abba9c97a0b3e3b5ddff6 |
| TPS | 64.77 | Not recorded | 698abba9c97a0b3e3b5ddff6 |
| R-TPT | 64.25 | Not recorded | 698abba9c97a0b3e3b5ddff6 |
| C-TPT | 63.85 | Not recorded | 698abba9c97a0b3e3b5ddff6 |
| RLCF | 63.64 | Not recorded | 698abba9c97a0b3e3b5ddff6 |
| TPT | 63.47 | Not recorded | 698abba9c97a0b3e3b5ddff6 |
| CLIP | 62.04 | Not recorded | 698abba9c97a0b3e3b5ddff6 |

### FCL (Ours)

Top-1 Acc (ImageNet): 65.55. Result ID: 699757fb5698665ec4aa1531.

[Reporting paper](<https://www.sota2.com/research/paper/698abba9c97a0b3e3b5ddff6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/698abba9c97a0b3e3b5ddff6.md>)

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

### ZERO

Top-1 Acc (ImageNet): 65.29. Result ID: 699757fb5698665ec4aa152d.

[Reporting paper](<https://www.sota2.com/research/paper/698abba9c97a0b3e3b5ddff6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75901e209131cea68a6d.md>)

[Implementation](<https://github.com/FarinaMatteo/zero>)

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

### MTA

Top-1 Acc (ImageNet): 65.02. Result ID: 699757fb5698665ec4aa152c.

[Reporting paper](<https://www.sota2.com/research/paper/698abba9c97a0b3e3b5ddff6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73191e209131cea5a896.md>)

[Implementation](<https://github.com/MaxZanella/MTA>)

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

### TTL

Top-1 Acc (ImageNet): 64.99. Result ID: 699757fb5698665ec4aa152e.

[Reporting paper](<https://www.sota2.com/research/paper/698abba9c97a0b3e3b5ddff6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa742e1e209131cea6048e.md>)

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

### TPS

Top-1 Acc (ImageNet): 64.77. Result ID: 699757fb5698665ec4aa152f.

[Reporting paper](<https://www.sota2.com/research/paper/698abba9c97a0b3e3b5ddff6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa762e1e209131cea6bb46.md>)

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

### R-TPT

Top-1 Acc (ImageNet): 64.25. Result ID: 699757fb5698665ec4aa1530.

[Reporting paper](<https://www.sota2.com/research/paper/698abba9c97a0b3e3b5ddff6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7add1e209131cea8590d.md>)

[Implementation](<https://github.com/TomSheng21/R-TPT>)

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

### C-TPT

Top-1 Acc (ImageNet): 63.85. Result ID: 699757fb5698665ec4aa152a.

[Reporting paper](<https://www.sota2.com/research/paper/698abba9c97a0b3e3b5ddff6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72bc1e209131cea58415.md>)

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

### RLCF

Top-1 Acc (ImageNet): 63.64. Result ID: 699757fb5698665ec4aa152b.

[Reporting paper](<https://www.sota2.com/research/paper/698abba9c97a0b3e3b5ddff6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72471e209131cea5581f.md>)

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

### TPT

Top-1 Acc (ImageNet): 63.47. Result ID: 699757fb5698665ec4aa1529.

[Reporting paper](<https://www.sota2.com/research/paper/698abba9c97a0b3e3b5ddff6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d691e209131cea399f9.md>)

[Implementation](<https://tinyurl.com/yr3zmhma>)

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

### CLIP

Top-1 Acc (ImageNet): 62.04. Result ID: 699757fb5698665ec4aa1528.

[Reporting paper](<https://www.sota2.com/research/paper/698abba9c97a0b3e3b5ddff6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69581e209131cea21ad0.md>)

[Implementation](<https://github.com/OpenAI/CLIP>)

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

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