# Image Classification on CIFAR-100 (50,000 labels) + Tiny Images (500,000 unlabeled)

Source: [Image Classification on CIFAR-100 (50,000 labels) + Tiny Images (500,000 unlabeled)](<https://www.sota2.com/research/sota/image-classification-on-cifar-100-50-000-labels-plus-tiny-images-500-000-unlabeled>)

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

Data updated: 2026-02-26T05:42:14.831Z

Evaluation of Image Classification performance using a combined dataset consisting of the 50,000 labeled examples from CIFAR-100 and 500,000 unlabeled images from Tiny Images.

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) | CIFAR-100 (50,000 labels) + Tiny Images (500,000 unlabeled) |
| Metric | Error Rate |
| Unit | PERCENT |
| Direction | Lower 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

- [Error Rate](<https://www.sota2.com/research/sota/image-classification-on-cifar-100-50-000-labels-plus-tiny-images-500-000-unlabeled.md?metric=Error%20Rate>): Lower is better; unit: PERCENT.

## Recorded results

| Method | Error Rate | Result date | Report paper |
| --- | --- | --- | --- |
| ImageNet-32 Fine-tuned | 22.22 | Not recorded | 68fa66ad1e209131cea11a49 |
| SESEMI ASL | 22.49 | Not recorded | 68fa66ad1e209131cea11a49 |
| SESEMI SSL | 22.52 | Not recorded | 68fa66ad1e209131cea11a49 |
| SESEMI SSL | 22.65 | Not recorded | 68fa66ad1e209131cea11a49 |
| TempEns SSL | 23.62 | Not recorded | 68fa66ad1e209131cea11a49 |
| II Model SSL | 25.79 | Not recorded | 68fa66ad1e209131cea11a49 |
| Supervised | 26.42 | Not recorded | 68fa66ad1e209131cea11a49 |

### ImageNet-32 Fine-tuned

Error Rate: 22.22. Result ID: 698d009a8b517eb4331e48d1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa66ad1e209131cea11a49.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa639f1e209131ce9ffcd4.md>)

[Implementation](<https://github.com/PatrykChrabaszcz/Imagenet32_Scripts>)

Recorded evaluation configuration: {"Learning Protocol":"Fine-tuned","Pre-trained":true}. This record does not assert independent reproduction.

### SESEMI ASL

Error Rate: 22.49. Result ID: 698d009a8b517eb4331e48d0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa66ad1e209131cea11a49.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66ad1e209131cea11a49.md>)

[Implementation](<https://github.com/vuptran/sesemi>)

Recorded evaluation configuration: {"Backbone":"ConvNet","Learning Protocol":"ASL"}. This record does not assert independent reproduction.

### SESEMI SSL

Error Rate: 22.52. Result ID: 698d009a8b517eb4331e48d4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa66ad1e209131cea11a49.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66ad1e209131cea11a49.md>)

[Implementation](<https://github.com/vuptran/sesemi>)

Recorded evaluation configuration: {"Backbone":"ConvNet","Learning Protocol":"SSL"}. This record does not assert independent reproduction.

### SESEMI SSL

Error Rate: 22.65. Result ID: 698d009a8b517eb4331e48d5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa66ad1e209131cea11a49.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66ad1e209131cea11a49.md>)

[Implementation](<https://github.com/vuptran/sesemi>)

Recorded evaluation configuration: {"Backbone":"WRN","Learning Protocol":"SSL"}. This record does not assert independent reproduction.

### TempEns SSL

Error Rate: 23.62. Result ID: 698d009a8b517eb4331e48d3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa66ad1e209131cea11a49.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63301e209131ce9fd2e9.md>)

Recorded evaluation configuration: {"Learning Protocol":"SSL"}. This record does not assert independent reproduction.

### II Model SSL

Error Rate: 25.79. Result ID: 698d009a8b517eb4331e48d2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa66ad1e209131cea11a49.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63301e209131ce9fd2e9.md>)

Recorded evaluation configuration: {"Learning Protocol":"SSL"}. This record does not assert independent reproduction.

### Supervised

Error Rate: 26.42. Result ID: 698d009a8b517eb4331e48cf.

[Reporting paper](<https://www.sota2.com/research/paper/68fa66ad1e209131cea11a49.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa64481e209131cea03b45.md>)

[Implementation](<https://github.com/CuriousAI/mean-teacher>)

Recorded evaluation configuration: {"Learning Protocol":"Supervised"}. This record does not assert independent reproduction.

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