# Image Classification on MNIST smallNORB MultiMNIST (test)

Source: [Image Classification on MNIST smallNORB MultiMNIST (test)](<https://www.sota2.com/research/sota/image-classification-on-mnist-smallnorb-multimnist-test>)

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

Data updated: 2026-02-26T05:57:16.487Z

Evaluation of Image Classification performance on the specialized MNIST smallNORB MultiMNIST test set, often including efficiency metrics like parameter count and computational operations.

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) | MNIST smallNORB MultiMNIST (test) |
| Metric | Parameters (K) |
| Unit | SIZE |
| Direction | Lower is better |
| Results loaded | 4 of 4 |
| 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

- [Parameters (K)](<https://www.sota2.com/research/sota/image-classification-on-mnist-smallnorb-multimnist-test.md?metric=Parameters%20(K)>): Lower is better; unit: SIZE.

- [OPS (G)](<https://www.sota2.com/research/sota/image-classification-on-mnist-smallnorb-multimnist-test.md?metric=OPS%20(G)>): Lower is better; unit: SIZE.

- [Improvement (%)](<https://www.sota2.com/research/sota/image-classification-on-mnist-smallnorb-multimnist-test.md?metric=Improvement%20(%25)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Parameters (K) | Result date | Report paper |
| --- | --- | --- | --- |
| Efficient-CapsNet | 161 | Not recorded | 68fa6b7e1e209131cea2e926 |
| Matrix-CapsNet with EM routing | 310 | Not recorded | 68fa6b7e1e209131cea2e926 |
| AR CapsNet | 5310 | Not recorded | 68fa6b7e1e209131cea2e926 |
| CapsNet | 6800 | Not recorded | 68fa6b7e1e209131cea2e926 |

### Efficient-CapsNet

Parameters (K): 161. Result ID: 698e8fdaffc3901de09eac32.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b7e1e209131cea2e926.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b7e1e209131cea2e926.md>)

[Implementation](<https://github.com/EscVM/Efficient-CapsNet>)

Recorded evaluation configuration: {"Batch Size":1}. This record does not assert independent reproduction.

### Matrix-CapsNet with EM routing

Parameters (K): 310. Result ID: 698e8fdaffc3901de09eac31.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b7e1e209131cea2e926.md>)

Recorded evaluation configuration: {"Batch Size":1}. This record does not assert independent reproduction.

### AR CapsNet

Parameters (K): 5310. Result ID: 698e8fdaffc3901de09eac30.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b7e1e209131cea2e926.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66ae1e209131cea11b75.md>)

Recorded evaluation configuration: {"Batch Size":1}. This record does not assert independent reproduction.

### CapsNet

Parameters (K): 6800. Result ID: 698e8fdaffc3901de09eac2f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b7e1e209131cea2e926.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63ce1e209131cea00bec.md>)

Recorded evaluation configuration: {"Batch Size":1}. This record does not assert independent reproduction.

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