# Image Classification on CIFAR-100-C Severity 5, Mild Scenario Source: [Image Classification on CIFAR-100-C Severity 5, Mild Scenario]() Collection index: [llms.txt]() Data updated: 2026-07-02T08:19:46.715Z Evaluation of Image Classification robustness on the CIFAR-100-C dataset under Severity Level 5 corruptions, specifically focusing on the 'Mild Scenario' configuration. Results report the error rate for 12 distinct corruption types (e.g., Gaussian Noise, Brightness) and the average error. 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-100-C Severity 5, Mild Scenario | | Metric | Gaussian Noise Error | | Unit | PERCENT | | Direction | Lower 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 - [Gaussian Noise Error](): Lower is better; unit: PERCENT. - [Shot Noise Error](): Lower is better; unit: PERCENT. - [Impulse Noise Error](): Lower is better; unit: PERCENT. - [Speckle Noise Error](): Lower is better; unit: PERCENT. - [Gaussian Blur Error](): Lower is better; unit: PERCENT. - [Brightness Error](): Lower is better; unit: PERCENT. - [Spatter Error](): Lower is better; unit: PERCENT. - [Contrast Error](): Lower is better; unit: PERCENT. - [Elastic Transform Error](): Lower is better; unit: PERCENT. - [Pixelate Error](): Lower is better; unit: PERCENT. - [JPEG Compression Error](): Lower is better; unit: PERCENT. - [Saturate Error](): Lower is better; unit: PERCENT. - [Average Error](): Lower is better; unit: PERCENT. ## Recorded results | Method | Gaussian Noise Error | Result date | Report paper | | --- | --- | --- | --- | | Tent | 1.93 | Not recorded | 6a45f082dc71372d7795a1b9 | | CLIP-ViT-B-32 | 5.8 | Not recorded | 6a45f082dc71372d7795a1b9 | | SVGD | 8.28 | Not recorded | 6a45f082dc71372d7795a1b9 | | KL | 9.31 | Not recorded | 6a45f082dc71372d7795a1b9 | | DeYO | 10.6 | Not recorded | 6a45f082dc71372d7795a1b9 | | Grad | 10.88 | Not recorded | 6a45f082dc71372d7795a1b9 | | Naive | 11.28 | Not recorded | 6a45f082dc71372d7795a1b9 | | SAR | 11.71 | Not recorded | 6a45f082dc71372d7795a1b9 | ### Tent Gaussian Noise Error: 1.93. Result ID: 6a461f2160b88bb33984dff8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Severity Level":"5","Batch Size":"64","Scenario":"Mild"}. This record does not assert independent reproduction. ### CLIP-ViT-B-32 Gaussian Noise Error: 5.8. Result ID: 6a461f2160b88bb33984dff7. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Severity Level":"5","Batch Size":"64","Scenario":"Mild"}. This record does not assert independent reproduction. ### SVGD Gaussian Noise Error: 8.28. Result ID: 6a461f2160b88bb33984dffc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Severity Level":"5","Batch Size":"64","Scenario":"Mild"}. This record does not assert independent reproduction. ### KL Gaussian Noise Error: 9.31. Result ID: 6a461f2160b88bb33984dffd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Severity Level":"5","Batch Size":"64","Scenario":"Mild"}. This record does not assert independent reproduction. ### DeYO Gaussian Noise Error: 10.6. Result ID: 6a461f2160b88bb33984dffa. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Severity Level":"5","Batch Size":"64","Scenario":"Mild"}. This record does not assert independent reproduction. ### Grad Gaussian Noise Error: 10.88. Result ID: 6a461f2160b88bb33984dffe. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Severity Level":"5","Batch Size":"64","Scenario":"Mild"}. This record does not assert independent reproduction. ### Naive Gaussian Noise Error: 11.28. Result ID: 6a461f2160b88bb33984dffb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Severity Level":"5","Batch Size":"64","Scenario":"Mild"}. This record does not assert independent reproduction. ### SAR Gaussian Noise Error: 11.71. Result ID: 6a461f2160b88bb33984dff9. [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.