# Image Classification on CIFAR-10-C Severity 5 Mild Scenario Source: [Image Classification on CIFAR-10-C Severity 5 Mild Scenario]() Collection index: [llms.txt]() Data updated: 2026-07-02T08:19:46.696Z Evaluation of Image Classification model robustness on the CIFAR-10-C dataset specifically configured at Severity Level 5 under mild scenario conditions, reporting performance across various common corruptions (e.g., noise, blur, compression). 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 5 Mild Scenario | | Metric | Accuracy (Gaussian Noise) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 8 of 8 | | 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 (Gaussian Noise)](): Higher is better; unit: PERCENT. - [Accuracy (Shot Noise)](): Higher is better; unit: PERCENT. - [Accuracy (Impulse Noise)](): Higher is better; unit: PERCENT. - [Accuracy (Speckle Noise)](): Higher is better; unit: PERCENT. - [Accuracy (Gaussian Blur)](): Higher is better; unit: PERCENT. - [Accuracy (Brightness)](): Higher is better; unit: PERCENT. - [Accuracy (Spatter)](): Higher is better; unit: PERCENT. - [Accuracy (Contrast)](): Higher is better; unit: PERCENT. - [Accuracy (Elastic Transform)](): Higher is better; unit: PERCENT. - [Accuracy (Pixelate)](): Higher is better; unit: PERCENT. - [Accuracy (JPEG Compression)](): Higher is better; unit: PERCENT. - [Accuracy (Saturate)](): Higher is better; unit: PERCENT. - [Average Accuracy (C5 Mild)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy (Gaussian Noise) | Result date | Report paper | | --- | --- | --- | --- | | Grad | 62.82 | Not recorded | 6a45f082dc71372d7795a1b9 | | Naive | 62.02 | Not recorded | 6a45f082dc71372d7795a1b9 | | SVGD | 61.28 | Not recorded | 6a45f082dc71372d7795a1b9 | | SAR | 41.4 | Not recorded | 6a45f082dc71372d7795a1b9 | | KL | 34.63 | Not recorded | 6a45f082dc71372d7795a1b9 | | DeYO | 33.35 | Not recorded | 6a45f082dc71372d7795a1b9 | | CLIP-ViT-B-32 | 25.21 | Not recorded | 6a45f082dc71372d7795a1b9 | | Tent | 13.54 | Not recorded | 6a45f082dc71372d7795a1b9 | ### Grad Accuracy (Gaussian Noise): 62.82. Result ID: 6a461f2160b88bb33984dff6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Severity Level":"5","Batch Size":"64","Scenario":"Mild"}. This record does not assert independent reproduction. ### Naive Accuracy (Gaussian Noise): 62.02. Result ID: 6a461f2160b88bb33984dff3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Severity Level":"5","Batch Size":"64","Scenario":"Mild"}. This record does not assert independent reproduction. ### SVGD Accuracy (Gaussian Noise): 61.28. Result ID: 6a461f2160b88bb33984dff4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Severity Level":"5","Batch Size":"64","Scenario":"Mild"}. This record does not assert independent reproduction. ### SAR Accuracy (Gaussian Noise): 41.4. Result ID: 6a461f2160b88bb33984dff1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Severity Level":"5","Batch Size":"64","Scenario":"Mild"}. This record does not assert independent reproduction. ### KL Accuracy (Gaussian Noise): 34.63. Result ID: 6a461f2160b88bb33984dff5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Severity Level":"5","Batch Size":"64","Scenario":"Mild"}. This record does not assert independent reproduction. ### DeYO Accuracy (Gaussian Noise): 33.35. Result ID: 6a461f2160b88bb33984dff2. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Severity Level":"5","Batch Size":"64","Scenario":"Mild"}. This record does not assert independent reproduction. ### CLIP-ViT-B-32 Accuracy (Gaussian Noise): 25.21. Result ID: 6a461f2160b88bb33984dfef. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Severity Level":"5","Batch Size":"64","Scenario":"Mild"}. This record does not assert independent reproduction. ### Tent Accuracy (Gaussian Noise): 13.54. Result ID: 6a461f2160b88bb33984dff0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Severity Level":"5","Batch Size":"64","Scenario":"Mild"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.