# Image Classification on Fashion-MNIST Class Unlearning (Drt) Source: [Image Classification on Fashion-MNIST Class Unlearning (Drt)]() Collection index: [llms.txt]() Data updated: 2026-05-12T08:04:04.993Z Utility evaluation (Accuracy) for models subjected to procedures on the Fashion-MNIST Class Unlearning (Drt) dataset variant. 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) | Fashion-MNIST Class Unlearning (Drt) | | 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 | | --- | --- | --- | --- | | Retrain | 96.71 | Not recorded | 68fa75ab1e209131cea69322 | | Finetune | 96.5 | Not recorded | 68fa75ab1e209131cea69322 | | SCRUB | 96.43 | Not recorded | 68fa75ab1e209131cea69322 | | Residual Feature Alignment Unlearning | 96.41 | Not recorded | 68fa75ab1e209131cea69322 | | BadT | 96.36 | Not recorded | 68fa75ab1e209131cea69322 | | Retrain | 96.33 | Not recorded | 68fa75ab1e209131cea69322 | | SCRUB | 96.07 | Not recorded | 68fa75ab1e209131cea69322 | | Residual Feature Alignment Unlearning | 95.94 | Not recorded | 68fa75ab1e209131cea69322 | | BadT | 95.77 | Not recorded | 68fa75ab1e209131cea69322 | | Finetune | 95.76 | Not recorded | 68fa75ab1e209131cea69322 | | Original | 95.47 | Not recorded | 68fa75ab1e209131cea69322 | | Original | 95.34 | Not recorded | 68fa75ab1e209131cea69322 | | NegGrad | 81.87 | Not recorded | 68fa75ab1e209131cea69322 | | NegGrad | 33.63 | Not recorded | 68fa75ab1e209131cea69322 | ### Retrain Accuracy: 96.71. Result ID: 6a02dee447f34ce924e57bd9. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"ViT"}. This record does not assert independent reproduction. ### Finetune Accuracy: 96.5. Result ID: 6a02dee447f34ce924e57bdb. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"ViT"}. This record does not assert independent reproduction. ### SCRUB Accuracy: 96.43. Result ID: 6a02dee447f34ce924e57bdd. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"ViT"}. This record does not assert independent reproduction. ### Residual Feature Alignment Unlearning Accuracy: 96.41. Result ID: 6a02dee447f34ce924e57bde. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ViT"}. This record does not assert independent reproduction. ### BadT Accuracy: 96.36. Result ID: 6a02dee447f34ce924e57bdc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ViT"}. This record does not assert independent reproduction. ### Retrain Accuracy: 96.33. Result ID: 6a02dee447f34ce924e57bd2. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"ResNet18"}. This record does not assert independent reproduction. ### SCRUB Accuracy: 96.07. Result ID: 6a02dee447f34ce924e57bd6. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"ResNet18"}. This record does not assert independent reproduction. ### Residual Feature Alignment Unlearning Accuracy: 95.94. Result ID: 6a02dee447f34ce924e57bd7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet18"}. This record does not assert independent reproduction. ### BadT Accuracy: 95.77. Result ID: 6a02dee447f34ce924e57bd5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet18"}. This record does not assert independent reproduction. ### Finetune Accuracy: 95.76. Result ID: 6a02dee447f34ce924e57bd4. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"ResNet18"}. This record does not assert independent reproduction. ### Original Accuracy: 95.47. Result ID: 6a02dee447f34ce924e57bd1. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"ResNet18"}. This record does not assert independent reproduction. ### Original Accuracy: 95.34. Result ID: 6a02dee447f34ce924e57bd8. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"ViT"}. This record does not assert independent reproduction. ### NegGrad Accuracy: 81.87. Result ID: 6a02dee447f34ce924e57bd3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet18"}. This record does not assert independent reproduction. ### NegGrad Accuracy: 33.63. Result ID: 6a02dee447f34ce924e57bda. [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.