# Image Classification on CIFAR-10-C Severity Level 3 (test) Source: [Image Classification on CIFAR-10-C Severity Level 3 (test)]() Collection index: [llms.txt]() Data updated: 2026-07-07T10:09:00.011Z Evaluation of Image Classification performance on the CIFAR-10-C dataset, specifically focusing on results obtained under Severity Level 3 corruptions across various types. 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) | CIFAR-10-C Severity Level 3 (test) | | Metric | Mean Test Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 5 of 5 | | 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 - [Mean Test Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Mean Test Accuracy | Result date | Report paper | | --- | --- | --- | --- | | FedAvg + SpecGradFilter | 77.68 | Not recorded | 6a4c882c85dba69fef8b2ad2 | | FedCM | 76.86 | Not recorded | 6a4c882c85dba69fef8b2ad2 | | FedDyn | 68.74 | Not recorded | 6a4c882c85dba69fef8b2ad2 | | FedAvg | 62.11 | Not recorded | 6a4c882c85dba69fef8b2ad2 | | SCAFFOLD | 61.25 | Not recorded | 6a4c882c85dba69fef8b2ad2 | ### FedAvg + SpecGradFilter Mean Test Accuracy: 77.68. Result ID: 6a4cd02760b88bb339ca4092. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-20","Severity Level":"3","Number of Clients":"100","Client Participation Ratio (R)":"0.1"}. This record does not assert independent reproduction. ### FedCM Mean Test Accuracy: 76.86. Result ID: 6a4cd02760b88bb339ca4090. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"ResNet-20","Severity Level":"3","Number of Clients":"100","Client Participation Ratio (R)":"0.1"}. This record does not assert independent reproduction. ### FedDyn Mean Test Accuracy: 68.74. Result ID: 6a4cd02760b88bb339ca408f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-20","Severity Level":"3","Number of Clients":"100","Client Participation Ratio (R)":"0.1"}. This record does not assert independent reproduction. ### FedAvg Mean Test Accuracy: 62.11. Result ID: 6a4cd02760b88bb339ca408e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-20","Severity Level":"3","Number of Clients":"100","Client Participation Ratio (R)":"0.1"}. This record does not assert independent reproduction. ### SCAFFOLD Mean Test Accuracy: 61.25. Result ID: 6a4cd02760b88bb339ca4091. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-20","Severity Level":"3","Number of Clients":"100","Client Participation Ratio (R)":"0.1"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.