# Machine Unlearning on CIFAR-10 forget split (10%) Source: [Machine Unlearning on CIFAR-10 forget split (10%)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:30:28.810Z Evaluation of Machine Unlearning effectiveness on the CIFAR-10 dataset, specifically utilizing a 10% subset designated as the forget set. Metrics include model accuracy and Membership Inference Attack (MIA) leakage. 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 | Machine Unlearning | | Dataset (including any stated split) | CIFAR-10 forget split (10%) | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 9 of 9 | | 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. - [MIA Leakage](): Lower is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | BadT | 100 | Not recorded | 68fa77cb1e209131cea75615 | | SalUn | 100 | Not recorded | 68fa77cb1e209131cea75615 | | ESC | 100 | Not recorded | 68fa77cb1e209131cea75615 | | SCRUB | 99.99 | Not recorded | 68fa77cb1e209131cea75615 | | ESC-T | 99.86 | Not recorded | 68fa77cb1e209131cea75615 | | Finetune | 99.85 | Not recorded | 68fa77cb1e209131cea75615 | | NG | 96.57 | Not recorded | 68fa77cb1e209131cea75615 | | RL | 94.52 | Not recorded | 68fa77cb1e209131cea75615 | | Retrain | 10 | Not recorded | 68fa77cb1e209131cea75615 | ### BadT Accuracy: 100. Result ID: 69956b233badfaffe65f4f9f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-18","Number of trials":3,"Run Time Efficiency (s)":182.78}. This record does not assert independent reproduction. ### SalUn Accuracy: 100. Result ID: 69956b233badfaffe65f4fa0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-18","Number of trials":3,"Run Time Efficiency (s)":150.51}. This record does not assert independent reproduction. ### ESC Accuracy: 100. Result ID: 69956b233badfaffe65f4fa2. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"ResNet-18","Number of trials":3,"Uses Retain set (Dr)":false,"Run Time Efficiency (s)":2.24}. This record does not assert independent reproduction. ### SCRUB Accuracy: 99.99. Result ID: 69956b233badfaffe65f4f9e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"ResNet-18","Number of trials":3,"Uses Retain set (Dr)":true,"Run Time Efficiency (s)":160.39}. This record does not assert independent reproduction. ### ESC-T Accuracy: 99.86. Result ID: 69956b233badfaffe65f4fa3. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"ResNet-18","Number of trials":3,"Uses Retain set (Dr)":false,"Run Time Efficiency (s)":29.47}. This record does not assert independent reproduction. ### Finetune Accuracy: 99.85. Result ID: 69956b233badfaffe65f4f9c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-18","Number of trials":3,"Run Time Efficiency (s)":132.02}. This record does not assert independent reproduction. ### NG Accuracy: 96.57. Result ID: 69956b233badfaffe65f4fa1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-18","Number of trials":3,"Run Time Efficiency (s)":26.79}. This record does not assert independent reproduction. ### RL Accuracy: 94.52. Result ID: 69956b233badfaffe65f4f9d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-18","Number of trials":3,"Run Time Efficiency (s)":2.57}. This record does not assert independent reproduction. ### Retrain Accuracy: 10. Result ID: 69956b233badfaffe65f4f9b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-18","Number of trials":3,"Run Time Efficiency (s)":2681.72}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.