# Image Classification on Digits MNIST SVHN MNIST-M USPS (test) Source: [Image Classification on Digits MNIST SVHN MNIST-M USPS (test)]() Collection index: [llms.txt]() Data updated: 2026-05-07T07:16:32.987Z Evaluation of Image Classification performance on the combined test split of the four digit datasets: MNIST, SVHN, MNIST-M, and USPS, reporting per-dataset scores and overall average. 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) | Digits MNIST SVHN MNIST-M USPS (test) | | Metric | Accuracy (MNIST) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 6 of 6 | | 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 (MNIST)](): Higher is better; unit: PERCENT. - [Accuracy (SVHN)](): Higher is better; unit: PERCENT. - [Accuracy (MNIST-M)](): Higher is better; unit: PERCENT. - [Accuracy (USPS)](): Higher is better; unit: PERCENT. - [Average Accuracy](): Higher is better; unit: PERCENT. - [Gain](): Higher is better; unit: SCALAR. ## Recorded results | Method | Accuracy (MNIST) | Result date | Report paper | | --- | --- | --- | --- | | Self-Distillation | 99.23 | Not recorded | 69fc1c7d5497b3b670c8845b | | Tbest (upper bound) | 99.2 | Not recorded | 69fc1c7d5497b3b670c8845b | | KL-divergence | 99.17 | Not recorded | 69fc1c7d5497b3b670c8845b | | LS | 99.13 | Not recorded | 69fc1c7d5497b3b670c8845b | | DKD | 98.7 | Not recorded | 69fc1c7d5497b3b670c8845b | | MDS | 98.15 | Not recorded | 69fc1c7d5497b3b670c8845b | ### Self-Distillation Accuracy (MNIST): 99.23. Result ID: 69fc3c4b47f34ce924de8ed7. [Reporting paper]() Recorded evaluation configuration: {"Internal Data":"MNIST"}. This record does not assert independent reproduction. ### Tbest (upper bound) Accuracy (MNIST): 99.2. Result ID: 69fc3c4b47f34ce924de8ed2. [Reporting paper]() Recorded evaluation configuration: {"Internal Data":"MNIST"}. This record does not assert independent reproduction. ### KL-divergence Accuracy (MNIST): 99.17. Result ID: 69fc3c4b47f34ce924de8ed3. [Reporting paper]() Recorded evaluation configuration: {"Internal Data":"MNIST"}. This record does not assert independent reproduction. ### LS Accuracy (MNIST): 99.13. Result ID: 69fc3c4b47f34ce924de8ed5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Internal Data":"MNIST"}. This record does not assert independent reproduction. ### DKD Accuracy (MNIST): 98.7. Result ID: 69fc3c4b47f34ce924de8ed4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Internal Data":"MNIST"}. This record does not assert independent reproduction. ### MDS Accuracy (MNIST): 98.15. Result ID: 69fc3c4b47f34ce924de8ed6. [Reporting paper]() Recorded evaluation configuration: {"Internal Data":"MNIST"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.