# Image Classification on MNIST (Certification and Rejection Metrics) Source: [Image Classification on MNIST (Certification and Rejection Metrics)]() Collection index: [llms.txt]() Data updated: 2026-06-29T06:50:24.931Z Evaluation of Image Classification performance on the MNIST dataset, reporting standard accuracy alongside certification, rejection, and radius 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) | MNIST | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 15 of 15 | | 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](): Higher is better; unit: PERCENT. - [Mean Radius](): Lower is better; unit: SCALAR. - [Average Score (All)](): Higher is better; unit: SCALAR. - [Average Certified Score](): Higher is better; unit: SCALAR. - [Average Rejected Score](): Higher is better; unit: SCALAR. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | KT | 99.2 | Not recorded | 6a41fc22206542251d7ccbdb | | Global Champ | 99.2 | Not recorded | 6a41fc22206542251d7ccbdb | | Specialist | 99.2 | Not recorded | 6a41fc22206542251d7ccbdb | | Cohen-10k | 99.2 | Not recorded | 6a41fc22206542251d7ccbdb | | Cohen-1k | 99.2 | Not recorded | 6a41fc22206542251d7ccbdb | | Cohen-10k | 98.8 | Not recorded | 6a41fc22206542251d7ccbdb | | Cohen-1k | 98.8 | Not recorded | 6a41fc22206542251d7ccbdb | | KT | 98.4 | Not recorded | 6a41fc22206542251d7ccbdb | | Global Champ | 98.4 | Not recorded | 6a41fc22206542251d7ccbdb | | Specialist | 98.4 | Not recorded | 6a41fc22206542251d7ccbdb | | Cohen-10k | 91.2 | Not recorded | 6a41fc22206542251d7ccbdb | | Cohen-1k | 89.6 | Not recorded | 6a41fc22206542251d7ccbdb | | KT | 89.2 | Not recorded | 6a41fc22206542251d7ccbdb | | Specialist | 89.2 | Not recorded | 6a41fc22206542251d7ccbdb | | Global Champ | 88.6 | Not recorded | 6a41fc22206542251d7ccbdb | ### KT Accuracy: 99.2. Result ID: 6a4215a9fbc373592c86b7dc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sigma":"0.25"}. This record does not assert independent reproduction. ### Global Champ Accuracy: 99.2. Result ID: 6a4215a9fbc373592c86b7dd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sigma":"0.25"}. This record does not assert independent reproduction. ### Specialist Accuracy: 99.2. Result ID: 6a4215a9fbc373592c86b7de. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sigma":"0.25"}. This record does not assert independent reproduction. ### Cohen-10k Accuracy: 99.2. Result ID: 6a4215a9fbc373592c86b7df. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sigma":"0.25"}. This record does not assert independent reproduction. ### Cohen-1k Accuracy: 99.2. Result ID: 6a4215a9fbc373592c86b7e0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sigma":"0.25"}. This record does not assert independent reproduction. ### Cohen-10k Accuracy: 98.8. Result ID: 6a4215a9fbc373592c86b7e4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sigma":"0.5"}. This record does not assert independent reproduction. ### Cohen-1k Accuracy: 98.8. Result ID: 6a4215a9fbc373592c86b7e5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sigma":"0.5"}. This record does not assert independent reproduction. ### KT Accuracy: 98.4. Result ID: 6a4215a9fbc373592c86b7e1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sigma":"0.5"}. This record does not assert independent reproduction. ### Global Champ Accuracy: 98.4. Result ID: 6a4215a9fbc373592c86b7e2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sigma":"0.5"}. This record does not assert independent reproduction. ### Specialist Accuracy: 98.4. Result ID: 6a4215a9fbc373592c86b7e3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sigma":"0.5"}. This record does not assert independent reproduction. ### Cohen-10k Accuracy: 91.2. Result ID: 6a4215a9fbc373592c86b7e9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sigma":"1."}. This record does not assert independent reproduction. ### Cohen-1k Accuracy: 89.6. Result ID: 6a4215a9fbc373592c86b7ea. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sigma":"1."}. This record does not assert independent reproduction. ### KT Accuracy: 89.2. Result ID: 6a4215a9fbc373592c86b7e6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sigma":"1."}. This record does not assert independent reproduction. ### Specialist Accuracy: 89.2. Result ID: 6a4215a9fbc373592c86b7e8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sigma":"1."}. This record does not assert independent reproduction. ### Global Champ Accuracy: 88.6. Result ID: 6a4215a9fbc373592c86b7e7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sigma":"1."}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.