# Subclass-level unlearning on CIFAR-100 (test) Source: [Subclass-level unlearning on CIFAR-100 (test)]() Collection index: [llms.txt]() Data updated: 2026-03-30T07:59:48.863Z Evaluation of Subclass-level Unlearning effectiveness on the test split of the CIFAR-100 dataset, reporting utility metrics such as Accuracy (Df), Accuracy (Dr-adj), and Accuracy (Dr-rem). 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 | Subclass-level unlearning | | Dataset (including any stated split) | CIFAR-100 (test) | | Metric | Accuracy (Df) | | 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 - [Accuracy (Df)](): Higher is better; unit: PERCENT. - [Accuracy (Dr-adj)](): Higher is better; unit: PERCENT. - [Accuracy (Dr-rem)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy (Df) | Result date | Report paper | | --- | --- | --- | --- | | Original | 90 | Not recorded | 69ca03a80636e5bf4afc949f | | FT | 62.33 | Not recorded | 69ca03a80636e5bf4afc949f | | GA | 56 | Not recorded | 69ca03a80636e5bf4afc949f | | ℓ1-sparse | 51.67 | Not recorded | 69ca03a80636e5bf4afc949f | | SSD | 33 | Not recorded | 69ca03a80636e5bf4afc949f | | Munba | 31.67 | Not recorded | 69ca03a80636e5bf4afc949f | | GDR | 8.67 | Not recorded | 69ca03a80636e5bf4afc949f | | SCRUB | 7 | Not recorded | 69ca03a80636e5bf4afc949f | | SalUn | 3 | Not recorded | 69ca03a80636e5bf4afc949f | | Our method | 2.33 | Not recorded | 69ca03a80636e5bf4afc949f | | DELETE | 0.67 | Not recorded | 69ca03a80636e5bf4afc949f | ### Original Accuracy (Df): 90. Result ID: 69ca2d5847f34ce924b2a05a. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"ResNet-18"}. This record does not assert independent reproduction. ### FT Accuracy (Df): 62.33. Result ID: 69ca2d5847f34ce924b2a05b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-18"}. This record does not assert independent reproduction. ### GA Accuracy (Df): 56. Result ID: 69ca2d5847f34ce924b2a05c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-18"}. This record does not assert independent reproduction. ### ℓ1-sparse Accuracy (Df): 51.67. Result ID: 69ca2d5847f34ce924b2a05d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-18"}. This record does not assert independent reproduction. ### SSD Accuracy (Df): 33. Result ID: 69ca2d5847f34ce924b2a05f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-18"}. This record does not assert independent reproduction. ### Munba Accuracy (Df): 31.67. Result ID: 69ca2d5847f34ce924b2a05e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-18"}. This record does not assert independent reproduction. ### GDR Accuracy (Df): 8.67. Result ID: 69ca2d5847f34ce924b2a063. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"ResNet-18"}. This record does not assert independent reproduction. ### SCRUB Accuracy (Df): 7. Result ID: 69ca2d5847f34ce924b2a060. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"ResNet-18"}. This record does not assert independent reproduction. ### SalUn Accuracy (Df): 3. Result ID: 69ca2d5847f34ce924b2a061. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-18"}. This record does not assert independent reproduction. ### Our method Accuracy (Df): 2.33. Result ID: 69ca2d5847f34ce924b2a064. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-18"}. This record does not assert independent reproduction. ### DELETE Accuracy (Df): 0.67. Result ID: 69ca2d5847f34ce924b2a062. [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.