# Image Classification on ImageNet-100-C Alternative Corruptions Source: [Image Classification on ImageNet-100-C Alternative Corruptions]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:35:12.268Z Evaluation of Image Classification performance on the ImageNet-100-C dataset variant, specifically measuring accuracy across various alternative corruption types (e.g., Brown, Caustic, Perlin). 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) | ImageNet-100-C Alternative Corruptions | | Metric | Average 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 - [Average Accuracy](): Higher is better; unit: PERCENT. - [Accuracy (BSmpl)](): Higher is better; unit: PERCENT. - [Accuracy (Brown Corruption)](): Higher is better; unit: PERCENT. - [Accuracy (Caustic Corruption)](): Higher is better; unit: PERCENT. - [Accuracy (Ckbd Corruption)](): Higher is better; unit: PERCENT. - [Accuracy (CSine Corruption)](): Higher is better; unit: PERCENT. - [Accuracy (ISpark Corruption)](): Higher is better; unit: PERCENT. - [Accuracy (Perlin Corruption)](): Higher is better; unit: PERCENT. - [Accuracy (Plasma Corruption)](): Higher is better; unit: PERCENT. - [Accuracy (SFreq Corruption)](): Higher is better; unit: PERCENT. - [Accuracy (Spark Corruption)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Average Accuracy | Result date | Report paper | | --- | --- | --- | --- | | PRIME | 65.9 | Not recorded | 68fa6d351e209131cea386e2 | | PRIME | 65 | Not recorded | 68fa6d351e209131cea386e2 | | DA | 63.8 | Not recorded | 68fa6d351e209131cea386e2 | | AugMix | 61 | Not recorded | 68fa6d351e209131cea386e2 | | Standard | 55.1 | Not recorded | 68fa6d351e209131cea386e2 | ### PRIME Average Accuracy: 65.9. Result ID: 698f83ed3a6c314849f99127. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"ResNet-18","Training Epochs":"1.5x"}. This record does not assert independent reproduction. ### PRIME Average Accuracy: 65. Result ID: 698f83ed3a6c314849f99126. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"ResNet-18"}. This record does not assert independent reproduction. ### DA Average Accuracy: 63.8. Result ID: 698f83ed3a6c314849f99125. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-18"}. This record does not assert independent reproduction. ### AugMix Average Accuracy: 61. Result ID: 698f83ed3a6c314849f99124. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-18"}. This record does not assert independent reproduction. ### Standard Average Accuracy: 55.1. Result ID: 698f83ed3a6c314849f99123. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-18"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.