# Image Classification on CIFAR-10 Class Unlearning (Df) Source: [Image Classification on CIFAR-10 Class Unlearning (Df)]() Collection index: [llms.txt]() Data updated: 2026-05-12T08:04:04.866Z Evaluation of standard Image Classification accuracy (utility) on the CIFAR-10 dataset variant specifically designed for Class Unlearning evaluation, denoted as (Df). 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 Class Unlearning (Df) | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 14 of 14 | | 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](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Original | 100 | Not recorded | 68fa75ab1e209131cea69322 | | Original | 100 | Not recorded | 68fa75ab1e209131cea69322 | | Finetune | 99.94 | Not recorded | 68fa75ab1e209131cea69322 | | Finetune | 94.22 | Not recorded | 68fa75ab1e209131cea69322 | | SCRUB | 67.64 | Not recorded | 68fa75ab1e209131cea69322 | | BadT | 15.02 | Not recorded | 68fa75ab1e209131cea69322 | | NegGrad | 12.42 | Not recorded | 68fa75ab1e209131cea69322 | | Residual Feature Alignment Unlearning | 0.42 | Not recorded | 68fa75ab1e209131cea69322 | | BadT | 0.06 | Not recorded | 68fa75ab1e209131cea69322 | | Retrain | 0 | Not recorded | 68fa75ab1e209131cea69322 | | Retrain | 0 | Not recorded | 68fa75ab1e209131cea69322 | | NegGrad | 0 | Not recorded | 68fa75ab1e209131cea69322 | | SCRUB | 0 | Not recorded | 68fa75ab1e209131cea69322 | | Residual Feature Alignment Unlearning | 0 | Not recorded | 68fa75ab1e209131cea69322 | ### Original Accuracy: 100. Result ID: 6a02dee447f34ce924e57b8b. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"ResNet18"}. This record does not assert independent reproduction. ### Original Accuracy: 100. Result ID: 6a02dee447f34ce924e57b92. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"ViT"}. This record does not assert independent reproduction. ### Finetune Accuracy: 99.94. Result ID: 6a02dee447f34ce924e57b95. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"ViT"}. This record does not assert independent reproduction. ### Finetune Accuracy: 94.22. Result ID: 6a02dee447f34ce924e57b8e. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"ResNet18"}. This record does not assert independent reproduction. ### SCRUB Accuracy: 67.64. Result ID: 6a02dee447f34ce924e57b90. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"ResNet18"}. This record does not assert independent reproduction. ### BadT Accuracy: 15.02. Result ID: 6a02dee447f34ce924e57b96. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ViT"}. This record does not assert independent reproduction. ### NegGrad Accuracy: 12.42. Result ID: 6a02dee447f34ce924e57b8d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet18"}. This record does not assert independent reproduction. ### Residual Feature Alignment Unlearning Accuracy: 0.42. Result ID: 6a02dee447f34ce924e57b91. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet18"}. This record does not assert independent reproduction. ### BadT Accuracy: 0.06. Result ID: 6a02dee447f34ce924e57b8f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet18"}. This record does not assert independent reproduction. ### Retrain Accuracy: 0. Result ID: 6a02dee447f34ce924e57b8c. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"ResNet18"}. This record does not assert independent reproduction. ### Retrain Accuracy: 0. Result ID: 6a02dee447f34ce924e57b93. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"ViT"}. This record does not assert independent reproduction. ### NegGrad Accuracy: 0. Result ID: 6a02dee447f34ce924e57b94. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ViT"}. This record does not assert independent reproduction. ### SCRUB Accuracy: 0. Result ID: 6a02dee447f34ce924e57b97. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"ViT"}. This record does not assert independent reproduction. ### Residual Feature Alignment Unlearning Accuracy: 0. Result ID: 6a02dee447f34ce924e57b98. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ViT"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.