# Image Classification on MNIST-R 64 rotations (test) Source: [Image Classification on MNIST-R 64 rotations (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T06:07:04.263Z Evaluation of Image Classification performance on the test split of the MNIST-R dataset, specifically testing robustness across 64 different rotation angles. 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-R 64 rotations (test) | | Metric | Top-1 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 - [Top-1 Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Top-1 Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Generalist | 98.9 | Not recorded | 68fa73731e209131cea5c98c | | ContextFlow++ | 97.9 | Not recorded | 68fa73731e209131cea5c98c | | ContextFlow++ | 97.9 | Not recorded | 68fa73731e209131cea5c98c | | Lu and Huang \[2020\] | 97.8 | Not recorded | 68fa73731e209131cea5c98c | | Lu and Huang \[2020\] | 97.8 | Not recorded | 68fa73731e209131cea5c98c | | ContextFlow++ | 97.8 | Not recorded | 68fa73731e209131cea5c98c | | ContextFlow++ | 97.8 | Not recorded | 68fa73731e209131cea5c98c | | Lu and Huang \[2020\] | 97.7 | Not recorded | 68fa73731e209131cea5c98c | | Lu and Huang \[2020\] | 97.7 | Not recorded | 68fa73731e209131cea5c98c | | ContextFlow++ | 97.7 | Not recorded | 68fa73731e209131cea5c98c | | Lu and Huang \[2020\] | 97.6 | Not recorded | 68fa73731e209131cea5c98c | | ContextFlow++ | 97.6 | Not recorded | 68fa73731e209131cea5c98c | | Lu and Huang \[2020\] | 97.4 | Not recorded | 68fa73731e209131cea5c98c | | Generalist | 96.1 | Not recorded | 68fa73731e209131cea5c98c | ### Generalist Top-1 Accuracy: 98.9. Result ID: 6992ed043a6c3148491267f8. [Reporting paper]() Recorded evaluation configuration: {"rotations":"without"}. This record does not assert independent reproduction. ### ContextFlow++ Top-1 Accuracy: 97.9. Result ID: 6992ed043a6c314849126803. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Fixed Generalist":true,"Context Encoder Category":"One-hot binary","Context Encoder Type":"variational"}. This record does not assert independent reproduction. ### ContextFlow++ Top-1 Accuracy: 97.9. Result ID: 6992ed043a6c314849126804. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Fixed Generalist":true,"Context Encoder Category":"Learned embedding","Context Encoder Type":"deterministic"}. This record does not assert independent reproduction. ### Lu and Huang \[2020\] Top-1 Accuracy: 97.8. Result ID: 6992ed043a6c3148491267fd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Fixed Generalist":false,"Context Encoder Category":"One-hot binary","Context Encoder Type":"variational"}. This record does not assert independent reproduction. ### Lu and Huang \[2020\] Top-1 Accuracy: 97.8. Result ID: 6992ed043a6c3148491267ff. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Fixed Generalist":false,"Context Encoder Category":"Learned embedding","Context Encoder Type":"stochastic"}. This record does not assert independent reproduction. ### ContextFlow++ Top-1 Accuracy: 97.8. Result ID: 6992ed043a6c314849126801. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Fixed Generalist":true,"Context Encoder Category":"Integer","Context Encoder Type":"arg max"}. This record does not assert independent reproduction. ### ContextFlow++ Top-1 Accuracy: 97.8. Result ID: 6992ed043a6c314849126805. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Fixed Generalist":true,"Context Encoder Category":"Learned embedding","Context Encoder Type":"stochastic"}. This record does not assert independent reproduction. ### Lu and Huang \[2020\] Top-1 Accuracy: 97.7. Result ID: 6992ed043a6c3148491267fb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Fixed Generalist":false,"Context Encoder Category":"Integer","Context Encoder Type":"arg max"}. This record does not assert independent reproduction. ### Lu and Huang \[2020\] Top-1 Accuracy: 97.7. Result ID: 6992ed043a6c3148491267fe. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Fixed Generalist":false,"Context Encoder Category":"Learned embedding","Context Encoder Type":"deterministic"}. This record does not assert independent reproduction. ### ContextFlow++ Top-1 Accuracy: 97.7. Result ID: 6992ed043a6c314849126800. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Fixed Generalist":true,"Context Encoder Category":"Integer","Context Encoder Type":"uniform"}. This record does not assert independent reproduction. ### Lu and Huang \[2020\] Top-1 Accuracy: 97.6. Result ID: 6992ed043a6c3148491267fa. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Fixed Generalist":false,"Context Encoder Category":"Integer","Context Encoder Type":"uniform"}. This record does not assert independent reproduction. ### ContextFlow++ Top-1 Accuracy: 97.6. Result ID: 6992ed043a6c314849126802. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Fixed Generalist":true,"Context Encoder Category":"One-hot binary","Context Encoder Type":"uniform"}. This record does not assert independent reproduction. ### Lu and Huang \[2020\] Top-1 Accuracy: 97.4. Result ID: 6992ed043a6c3148491267fc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Fixed Generalist":false,"Context Encoder Category":"One-hot binary","Context Encoder Type":"uniform"}. This record does not assert independent reproduction. ### Generalist Top-1 Accuracy: 96.1. Result ID: 6992ed043a6c3148491267f9. [Reporting paper]() Recorded evaluation configuration: {"rotations":"with"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.