# Image Classification on CIFAR-10 (test) (Uniform Noise and Flip Perturbations) Source: [Image Classification on CIFAR-10 (test) (Uniform Noise and Flip Perturbations)]() Collection index: [llms.txt]() Data updated: 2026-05-19T11:42:37.444Z Evaluation of Image Classification model performance on the CIFAR-10 test set under varying levels of uniform input noise corruption (0% to 80%) and specific flip perturbations, also reporting average rank. 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 (test) | | Metric | Accuracy (Uniform Noise 0%) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 9 of 9 | | 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 (Uniform Noise 0%)](): Higher is better; unit: PERCENT. - [Accuracy (Uniform Noise 20%)](): Higher is better; unit: PERCENT. - [Accuracy (Uniform Noise 40%)](): Higher is better; unit: PERCENT. - [Accuracy (Uniform Noise 60%)](): Higher is better; unit: PERCENT. - [Accuracy (Uniform Noise 80%)](): Higher is better; unit: PERCENT. - [Accuracy (Flip Perturbation 20%)](): Higher is better; unit: PERCENT. - [Accuracy (Flip Perturbation 40%)](): Higher is better; unit: PERCENT. - [Average Rank](): Lower is better; unit: SCALAR. ## Recorded results | Method | Accuracy (Uniform Noise 0%) | Result date | Report paper | | --- | --- | --- | --- | | EBOMLC | 92.92 | Not recorded | 6a0beef1dbc836b87d82c8fb | | MLC | 92.76 | Not recorded | 6a0beef1dbc836b87d82c8fb | | GLC | 92.25 | Not recorded | 6a0beef1dbc836b87d82c8fb | | DivideMix | 91.32 | Not recorded | 6a0beef1dbc836b87d82c8fb | | PMW-Net | 90.74 | Not recorded | 6a0beef1dbc836b87d82c8fb | | Co-teaching+ | 89.14 | Not recorded | 6a0beef1dbc836b87d82c8fb | | MW-Net | 88.91 | Not recorded | 6a0beef1dbc836b87d82c8fb | | L2RW | 85.16 | Not recorded | 6a0beef1dbc836b87d82c8fb | | MLC-D | 67.75 | Not recorded | 6a0beef1dbc836b87d82c8fb | ### EBOMLC Accuracy (Uniform Noise 0%): 92.92. Result ID: 6a0c4c20fbc373592c036f86. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-32 (frozen)","Training Time":"1.00×"}. This record does not assert independent reproduction. ### MLC Accuracy (Uniform Noise 0%): 92.76. Result ID: 6a0c4c20fbc373592c036f84. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"ResNet-32 (frozen)","Training Time":"2.28×"}. This record does not assert independent reproduction. ### GLC Accuracy (Uniform Noise 0%): 92.25. Result ID: 6a0c4c20fbc373592c036f80. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-32","Training Time":"0.33×"}. This record does not assert independent reproduction. ### DivideMix Accuracy (Uniform Noise 0%): 91.32. Result ID: 6a0c4c20fbc373592c036f7f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"ResNet-32","Training Time":"1.12×"}. This record does not assert independent reproduction. ### PMW-Net Accuracy (Uniform Noise 0%): 90.74. Result ID: 6a0c4c20fbc373592c036f83. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"ResNet-32","Training Time":"1.95×"}. This record does not assert independent reproduction. ### Co-teaching+ Accuracy (Uniform Noise 0%): 89.14. Result ID: 6a0c4c20fbc373592c036f7e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-32","Training Time":"0.51×"}. This record does not assert independent reproduction. ### MW-Net Accuracy (Uniform Noise 0%): 88.91. Result ID: 6a0c4c20fbc373592c036f82. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"ResNet-32","Training Time":"1.89×"}. This record does not assert independent reproduction. ### L2RW Accuracy (Uniform Noise 0%): 85.16. Result ID: 6a0c4c20fbc373592c036f81. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-32","Training Time":"1.69×"}. This record does not assert independent reproduction. ### MLC-D Accuracy (Uniform Noise 0%): 67.75. Result ID: 6a0c4c20fbc373592c036f85. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"ResNet-32 (frozen)","k":"5","Training Time":"2.32×"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.