# Image Classification on EMNIST Public Task Finetuned (test)

Source: [Image Classification on EMNIST Public Task Finetuned (test)](<https://www.sota2.com/research/sota/image-classification-on-emnist-public-task-finetuned-test>)

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

Data updated: 2026-03-05T08:07:41.818Z

Evaluation of Image Classification performance on the test split of the EMNIST Public Task Finetuned configuration.

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) | EMNIST Public Task Finetuned (test) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 14 of 14 |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/image-classification-on-emnist-public-task-finetuned-test.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Quantum Augmentation Baseline | 88.5 | Not recorded | 68fa43ba1e209131ce90c1f0 |
| Gaussian Blur Augmented EMNIST Baseline | 83.9 | Not recorded | 68fa43ba1e209131ce90c1f0 |
| Accuracy-Based Reward Policy | 64.4 | Not recorded | 68fa43ba1e209131ce90c1f0 |
| Length-Based Reward Policy | 63.9 | Not recorded | 68fa43ba1e209131ce90c1f0 |
| Gaussian Noise Augmented EMNIST Baseline | 56.5 | Not recorded | 68fa43ba1e209131ce90c1f0 |
| Public-Based Reward Policy | 56.5 | Not recorded | 68fa43ba1e209131ce90c1f0 |
| Gaussian Blur Augmented EMNIST Baseline | 54.1 | Not recorded | 68fa43ba1e209131ce90c1f0 |
| Quantum Random Action Baseline | 51.9 | Not recorded | 68fa43ba1e209131ce90c1f0 |
| Gaussian Noise Augmented EMNIST Baseline | 51.5 | Not recorded | 68fa43ba1e209131ce90c1f0 |
| Length-Based Reward Policy | 51.5 | Not recorded | 68fa43ba1e209131ce90c1f0 |
| Public-Based Reward Policy | 50.7 | Not recorded | 68fa43ba1e209131ce90c1f0 |
| Pure Chance Baseline | 50 | Not recorded | 68fa43ba1e209131ce90c1f0 |
| Quantum Random Action Baseline | 49.6 | Not recorded | 68fa43ba1e209131ce90c1f0 |
| Accuracy-Based Reward Policy | 48.7 | Not recorded | 68fa43ba1e209131ce90c1f0 |

### Quantum Augmentation Baseline

Accuracy: 88.5. Result ID: 69a939956dbb7075b9527a33.

[Reporting paper](<https://www.sota2.com/research/paper/68fa43ba1e209131ce90c1f0.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Gaussian Blur Augmented EMNIST Baseline

Accuracy: 83.9. Result ID: 69a939956dbb7075b9527a37.

[Reporting paper](<https://www.sota2.com/research/paper/68fa43ba1e209131ce90c1f0.md>)

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

### Accuracy-Based Reward Policy

Accuracy: 64.4. Result ID: 69a939956dbb7075b9527a3e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa43ba1e209131ce90c1f0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa43ba1e209131ce90c1f0.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Testing","Training Steps":"52,829","Training Hours":"348"}. This record does not assert independent reproduction.

### Length-Based Reward Policy

Accuracy: 63.9. Result ID: 69a939956dbb7075b9527a3c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa43ba1e209131ce90c1f0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa43ba1e209131ce90c1f0.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Testing","Training Steps":"28,519","Training Hours":"190"}. This record does not assert independent reproduction.

### Gaussian Noise Augmented EMNIST Baseline

Accuracy: 56.5. Result ID: 69a939956dbb7075b9527a39.

[Reporting paper](<https://www.sota2.com/research/paper/68fa43ba1e209131ce90c1f0.md>)

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

### Public-Based Reward Policy

Accuracy: 56.5. Result ID: 69a939956dbb7075b9527a3b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa43ba1e209131ce90c1f0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa43ba1e209131ce90c1f0.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Finetuned","Training Steps":"90,397","Training Hours":"605"}. This record does not assert independent reproduction.

### Gaussian Blur Augmented EMNIST Baseline

Accuracy: 54.1. Result ID: 69a939956dbb7075b9527a36.

[Reporting paper](<https://www.sota2.com/research/paper/68fa43ba1e209131ce90c1f0.md>)

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

### Quantum Random Action Baseline

Accuracy: 51.9. Result ID: 69a939956dbb7075b9527a35.

[Reporting paper](<https://www.sota2.com/research/paper/68fa43ba1e209131ce90c1f0.md>)

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

### Gaussian Noise Augmented EMNIST Baseline

Accuracy: 51.5. Result ID: 69a939956dbb7075b9527a38.

[Reporting paper](<https://www.sota2.com/research/paper/68fa43ba1e209131ce90c1f0.md>)

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

### Length-Based Reward Policy

Accuracy: 51.5. Result ID: 69a939956dbb7075b9527a3d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa43ba1e209131ce90c1f0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa43ba1e209131ce90c1f0.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Finetuned","Training Steps":"28,519","Training Hours":"190"}. This record does not assert independent reproduction.

### Public-Based Reward Policy

Accuracy: 50.7. Result ID: 69a939956dbb7075b9527a3a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa43ba1e209131ce90c1f0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa43ba1e209131ce90c1f0.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Testing","Training Steps":"90,397","Training Hours":"605"}. This record does not assert independent reproduction.

### Pure Chance Baseline

Accuracy: 50. Result ID: 69a939956dbb7075b9527a32.

[Reporting paper](<https://www.sota2.com/research/paper/68fa43ba1e209131ce90c1f0.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Quantum Random Action Baseline

Accuracy: 49.6. Result ID: 69a939956dbb7075b9527a34.

[Reporting paper](<https://www.sota2.com/research/paper/68fa43ba1e209131ce90c1f0.md>)

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

### Accuracy-Based Reward Policy

Accuracy: 48.7. Result ID: 69a939956dbb7075b9527a3f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa43ba1e209131ce90c1f0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa43ba1e209131ce90c1f0.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"Finetuned","Training Steps":"52,829","Training Hours":"348"}. This record does not assert independent reproduction.

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