# Image Classification on ImageNet (Standard and Robust Accuracy) Source: [Image Classification on ImageNet (Standard and Robust Accuracy)]() Collection index: [llms.txt]() Data updated: 2026-02-27T10:14:31.758Z Evaluation of Image Classification performance on the standard ImageNet dataset, reporting both standard accuracy and robustness accuracy scores. 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 | | Metric | Standard Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 11 of 11 | | 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 - [Standard Accuracy](): Higher is better; unit: PERCENT. - [Robust Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Standard Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Without defense | 82.33 | Not recorded | 68fa77991e209131cea74467 | | Without defense | 80.55 | Not recorded | 68fa77991e209131cea74467 | | Without defense | 80.55 | Not recorded | 68fa77991e209131cea74467 | | OSCP | 79.81 | Not recorded | 68fa77991e209131cea74467 | | Wang et al. | 78.1 | Not recorded | 68fa77991e209131cea74467 | | Amini et al. | 77.96 | Not recorded | 68fa77991e209131cea74467 | | OSCP | 77.63 | Not recorded | 68fa77991e209131cea74467 | | OSCP | 77.63 | Not recorded | 68fa77991e209131cea74467 | | Singh et al. | 77 | Not recorded | 68fa77991e209131cea74467 | | DiffPure | 75.77 | Not recorded | 68fa77991e209131cea74467 | | Wang et al. | 73.53 | Not recorded | 68fa77991e209131cea74467 | ### Without defense Standard Accuracy: 82.33. Result ID: 69a16e6e35dfcabb00e70104. [Reporting paper]() Recorded evaluation configuration: {"Attack Method":"random-targeted PGD-40","Architecture":"ResNet-152"}. This record does not assert independent reproduction. ### Without defense Standard Accuracy: 80.55. Result ID: 69a16e6e35dfcabb00e70102. [Reporting paper]() Recorded evaluation configuration: {"Attack Method":"untargeted PGD-100","Architecture":"ResNet-50"}. This record does not assert independent reproduction. ### Without defense Standard Accuracy: 80.55. Result ID: 69a16e6e35dfcabb00e70103. [Reporting paper]() Recorded evaluation configuration: {"Attack Method":"AutoAttack","Architecture":"ResNet-50"}. This record does not assert independent reproduction. ### OSCP Standard Accuracy: 79.81. Result ID: 69a16e6e35dfcabb00e7010c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Defense":"Hybrid","Attack Method":"random-targeted PGD-40","Architecture":"ResNet-152"}. This record does not assert independent reproduction. ### Wang et al. Standard Accuracy: 78.1. Result ID: 69a16e6e35dfcabb00e70106. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Defense":"Diffusion Based Purification","Attack Method":"random-targeted PGD-40","Architecture":"ResNet-152"}. This record does not assert independent reproduction. ### Amini et al. Standard Accuracy: 77.96. Result ID: 69a16e6e35dfcabb00e70108. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Defense":"Adversarial Training","Attack Method":"AutoAttack","Architecture":"ConvNeXt-L"}. This record does not assert independent reproduction. ### OSCP Standard Accuracy: 77.63. Result ID: 69a16e6e35dfcabb00e7010a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Defense":"Hybrid","Attack Method":"untargeted PGD-100","Architecture":"ResNet-50"}. This record does not assert independent reproduction. ### OSCP Standard Accuracy: 77.63. Result ID: 69a16e6e35dfcabb00e7010b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Defense":"Hybrid","Attack Method":"AutoAttack","Architecture":"ResNet-50"}. This record does not assert independent reproduction. ### Singh et al. Standard Accuracy: 77. Result ID: 69a16e6e35dfcabb00e70109. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Defense":"Adversarial Training","Attack Method":"AutoAttack","Architecture":"ConvNeXt-L"}. This record does not assert independent reproduction. ### DiffPure Standard Accuracy: 75.77. Result ID: 69a16e6e35dfcabb00e70107. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Defense":"Diffusion Based Purification","Attack Method":"AutoAttack","Architecture":"ResNet-50"}. This record does not assert independent reproduction. ### Wang et al. Standard Accuracy: 73.53. Result ID: 69a16e6e35dfcabb00e70105. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Defense":"Diffusion Based Purification","Attack Method":"untargeted PGD-100","Architecture":"ResNet-50"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.