# Image Classification on ImageNet 1k (val) (Robustness Accuracy & CRA) Source: [Image Classification on ImageNet 1k (val) (Robustness Accuracy & CRA)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:39:21.103Z Evaluation of Image Classification robustness on the ImageNet 1k validation set, reporting Clean Accuracy and Certified Robustness Accuracy (CRA) under specified perturbation budgets (e.g., epsilon=1 and epsilon=36/255). 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 1k (val) | | Metric | Clean Accuracy (eps=36/255) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 6 of 6 | | 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 - [Clean Accuracy (eps=36/255)](): Higher is better; unit: PERCENT. - [Clean Accuracy (eps=1)](): Higher is better; unit: PERCENT. - [CRA (eps=1)](): Higher is better; unit: PERCENT. - [CRA (eps=36/255)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Clean Accuracy (eps=36/255) | Result date | Report paper | | --- | --- | --- | --- | | LipNeXt L32W5792 | 57 | Not recorded | 697846dfa2b28e761d87c821 | | LipNeXt L32W4096 | 55.9 | Not recorded | 697846dfa2b28e761d87c821 | | BRONet | 52.3 | Not recorded | 697846dfa2b28e761d87c821 | | BRONet | 49.3 | Not recorded | 697846dfa2b28e761d87c821 | | LiResNet++ | 49 | Not recorded | 697846dfa2b28e761d87c821 | | LiResNet | 45.6 | Not recorded | 697846dfa2b28e761d87c821 | ### LipNeXt L32W5792 Clean Accuracy (eps=36/255): 57. Result ID: 6999158b5698665ec47bae9e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"#Param.":"2B","Training Speed":"17.8","Use EDM":false}. This record does not assert independent reproduction. ### LipNeXt L32W4096 Clean Accuracy (eps=36/255): 55.9. Result ID: 6999158b5698665ec47bae9d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"#Param.":"1B","Training Speed":"8.9","Use EDM":false}. This record does not assert independent reproduction. ### BRONet Clean Accuracy (eps=36/255): 52.3. Result ID: 6999158b5698665ec47bae9c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"#Param.":"86M","Use EDM":true}. This record does not assert independent reproduction. ### BRONet Clean Accuracy (eps=36/255): 49.3. Result ID: 6999158b5698665ec47bae9b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"#Param.":"86M","Training Speed":"10.5","Use EDM":false}. This record does not assert independent reproduction. ### LiResNet++ Clean Accuracy (eps=36/255): 49. Result ID: 6999158b5698665ec47bae9a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"#Param.":"83M","Use EDM":true}. This record does not assert independent reproduction. ### LiResNet Clean Accuracy (eps=36/255): 45.6. Result ID: 6999158b5698665ec47bae99. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"#Param.":"51M","Training Speed":"5.3","Use EDM":false}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.