# Image Classification on ImageNet original (val) (Inductive/Transductive Accuracy)

Source: [Image Classification on ImageNet original (val) (Inductive/Transductive Accuracy)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-original-val-inductive-transductive-accuracy>)

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

Data updated: 2026-02-26T05:47:06.280Z

Evaluation of Image Classification performance on the validation split of the original ImageNet dataset, specifically reporting Inductive and Transductive Top-1 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) | ImageNet original (val) |
| Metric | Inductive Top-1 Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 17 of 17 |
| 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

- [Inductive Top-1 Accuracy](<https://www.sota2.com/research/sota/image-classification-on-imagenet-original-val-inductive-transductive-accuracy.md?metric=Inductive%20Top-1%20Accuracy>): Higher is better; unit: PERCENT.

- [Transductive Top-1 Accuracy](<https://www.sota2.com/research/sota/image-classification-on-imagenet-original-val-inductive-transductive-accuracy.md?metric=Transductive%20Top-1%20Accuracy>): Higher is better; unit: PERCENT.

- [Improvement Percentage](<https://www.sota2.com/research/sota/image-classification-on-imagenet-original-val-inductive-transductive-accuracy.md?metric=Improvement%20Percentage>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Inductive Top-1 Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| TransBoost | 82.05 | Not recorded | 68fa6e5a1e209131cea3f4dd |
| TransBoost | 81.39 | Not recorded | 68fa6e5a1e209131cea3f4dd |
| TransBoost | 81.38 | Not recorded | 68fa6e5a1e209131cea3f4dd |
| TransBoost | 80.38 | Not recorded | 68fa6e5a1e209131cea3f4dd |
| TransBoost | 78.33 | Not recorded | 68fa6e5a1e209131cea3f4dd |
| SimCLR | 78.33 | Not recorded | 68fa6e5a1e209131cea3f4dd |
| SimCLRv2 | 78.33 | Not recorded | 68fa6e5a1e209131cea3f4dd |
| TransBoost | 77.7 | Not recorded | 68fa6e5a1e209131cea3f4dd |
| TransBoost | 77.37 | Not recorded | 68fa6e5a1e209131cea3f4dd |
| TransBoost | 76.15 | Not recorded | 68fa6e5a1e209131cea3f4dd |
| SimCLR | 76.15 | Not recorded | 68fa6e5a1e209131cea3f4dd |
| SimCLRv2 | 76.15 | Not recorded | 68fa6e5a1e209131cea3f4dd |
| Ent-min | 76.15 | Not recorded | 68fa6e5a1e209131cea3f4dd |
| TIM-GD | 76.15 | Not recorded | 68fa6e5a1e209131cea3f4dd |
| TransBoost | 75.78 | Not recorded | 68fa6e5a1e209131cea3f4dd |
| TransBoost | 73.29 | Not recorded | 68fa6e5a1e209131cea3f4dd |
| TransBoost | 69.76 | Not recorded | 68fa6e5a1e209131cea3f4dd |

### TransBoost

Inductive Top-1 Accuracy: 82.05. Result ID: 699015413a6c314849fe00b4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4dd.md>)

[Implementation](<https://github.com/omerb01/TransBoost>)

Recorded evaluation configuration: {"Architecture":"ConvNext-T","Params (M)":"28.59","Pre-training Procedure":"Advanced procedure"}. This record does not assert independent reproduction.

### TransBoost

Inductive Top-1 Accuracy: 81.39. Result ID: 699015413a6c314849fe00b5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4dd.md>)

[Implementation](<https://github.com/omerb01/TransBoost>)

Recorded evaluation configuration: {"Architecture":"ViT-S","Params (M)":"22.05","Pre-training Procedure":"Advanced procedure"}. This record does not assert independent reproduction.

### TransBoost

Inductive Top-1 Accuracy: 81.38. Result ID: 699015413a6c314849fe00b6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4dd.md>)

[Implementation](<https://github.com/omerb01/TransBoost>)

Recorded evaluation configuration: {"Architecture":"Swin-T","Params (M)":"28.29","Pre-training Procedure":"Advanced procedure"}. This record does not assert independent reproduction.

### TransBoost

Inductive Top-1 Accuracy: 80.38. Result ID: 699015413a6c314849fe00b3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4dd.md>)

[Implementation](<https://github.com/omerb01/TransBoost>)

Recorded evaluation configuration: {"Architecture":"ResNet50-StrikesBack","Params (M)":"25.56","Pre-training Procedure":"Advanced procedure"}. This record does not assert independent reproduction.

### TransBoost

Inductive Top-1 Accuracy: 78.33. Result ID: 699015413a6c314849fe00b0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4dd.md>)

[Implementation](<https://github.com/omerb01/TransBoost>)

Recorded evaluation configuration: {"Architecture":"ResNet152","Params (M)":"60.19","Pre-training Procedure":"Standard simple procedure"}. This record does not assert independent reproduction.

### SimCLR

Inductive Top-1 Accuracy: 78.33. Result ID: 699015413a6c314849fe00b8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67d71e209131cea18de5.md>)

[Implementation](<https://github.com/google-research/simclr>)

Recorded evaluation configuration: {"Architecture":"ResNet152","Params (M)":"60.19","Learning Mode":"Semi-supervised learning"}. This record does not assert independent reproduction.

### SimCLRv2

Inductive Top-1 Accuracy: 78.33. Result ID: 699015413a6c314849fe00ba.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68a21e209131cea1d4d1.md>)

[Implementation](<https://github.com/google-research/simclr>)

Recorded evaluation configuration: {"Architecture":"ResNet152","Params (M)":"60.19","Learning Mode":"Semi-supervised learning"}. This record does not assert independent reproduction.

### TransBoost

Inductive Top-1 Accuracy: 77.7. Result ID: 699015413a6c314849fe00b1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4dd.md>)

[Implementation](<https://github.com/omerb01/TransBoost>)

Recorded evaluation configuration: {"Architecture":"EfficientNetB0","Params (M)":"05.29","Pre-training Procedure":"Advanced procedure"}. This record does not assert independent reproduction.

### TransBoost

Inductive Top-1 Accuracy: 77.37. Result ID: 699015413a6c314849fe00af.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4dd.md>)

[Implementation](<https://github.com/omerb01/TransBoost>)

Recorded evaluation configuration: {"Architecture":"ResNet101","Params (M)":"44.55","Pre-training Procedure":"Standard simple procedure"}. This record does not assert independent reproduction.

### TransBoost

Inductive Top-1 Accuracy: 76.15. Result ID: 699015413a6c314849fe00ae.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4dd.md>)

[Implementation](<https://github.com/omerb01/TransBoost>)

Recorded evaluation configuration: {"Architecture":"ResNet50","Params (M)":"25.56","Pre-training Procedure":"Standard simple procedure"}. This record does not assert independent reproduction.

### SimCLR

Inductive Top-1 Accuracy: 76.15. Result ID: 699015413a6c314849fe00b7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67d71e209131cea18de5.md>)

[Implementation](<https://github.com/google-research/simclr>)

Recorded evaluation configuration: {"Architecture":"ResNet50","Params (M)":"25.56","Learning Mode":"Semi-supervised learning"}. This record does not assert independent reproduction.

### SimCLRv2

Inductive Top-1 Accuracy: 76.15. Result ID: 699015413a6c314849fe00b9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68a21e209131cea1d4d1.md>)

[Implementation](<https://github.com/google-research/simclr>)

Recorded evaluation configuration: {"Architecture":"ResNet50","Params (M)":"25.56","Learning Mode":"Semi-supervised learning"}. This record does not assert independent reproduction.

### Ent-min

Inductive Top-1 Accuracy: 76.15. Result ID: 699015413a6c314849fe00bb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68891e209131cea1cc20.md>)

Recorded evaluation configuration: {"Architecture":"ResNet50","Params (M)":"25.56","Learning Mode":"Transductive few-shot learning"}. This record does not assert independent reproduction.

### TIM-GD

Inductive Top-1 Accuracy: 76.15. Result ID: 699015413a6c314849fe00bc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68a01e209131cea1d398.md>)

[Implementation](<https://github.com/mboudiaf/TIM>)

Recorded evaluation configuration: {"Architecture":"ResNet50","Params (M)":"25.56","Learning Mode":"Transductive few-shot learning"}. This record does not assert independent reproduction.

### TransBoost

Inductive Top-1 Accuracy: 75.78. Result ID: 699015413a6c314849fe00b2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4dd.md>)

[Implementation](<https://github.com/omerb01/TransBoost>)

Recorded evaluation configuration: {"Architecture":"MobileNetV3-L","Params (M)":"05.48","Pre-training Procedure":"Advanced procedure"}. This record does not assert independent reproduction.

### TransBoost

Inductive Top-1 Accuracy: 73.29. Result ID: 699015413a6c314849fe00ad.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4dd.md>)

[Implementation](<https://github.com/omerb01/TransBoost>)

Recorded evaluation configuration: {"Architecture":"ResNet34","Params (M)":"21.80","Pre-training Procedure":"Standard simple procedure"}. This record does not assert independent reproduction.

### TransBoost

Inductive Top-1 Accuracy: 69.76. Result ID: 699015413a6c314849fe00ac.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e5a1e209131cea3f4dd.md>)

[Implementation](<https://github.com/omerb01/TransBoost>)

Recorded evaluation configuration: {"Architecture":"ResNet18","Params (M)":"11.69","Pre-training Procedure":"Standard simple procedure"}. This record does not assert independent reproduction.

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