# Machine Unlearning Effectiveness Assessment on CIFAR-10 (UA/MIA Focus) Source: [Machine Unlearning Effectiveness Assessment on CIFAR-10 (UA/MIA Focus)]() Collection index: [llms.txt]() Data updated: 2026-05-13T10:24:29.684Z Evaluation of Machine Unlearning effectiveness on CIFAR-10, focusing on metrics quantifying unlearning accuracy (UA, RA), privacy leakage assessment via Membership Inference Attacks (MIA), and residual error metrics (RTE, TA). 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 | | Metric | Unlearning Accuracy (UA) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 10 of 10 | | 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 - [Unlearning Accuracy (UA)](): Higher is better; unit: PERCENT. - [Test Unlearning Accuracy (UAtest)](): Higher is better; unit: PERCENT. - [Retained Accuracy (RA)](): Higher is better; unit: PERCENT. - [Target Accuracy (TA)](): Higher is better; unit: PERCENT. - [MIA Score](): Lower is better; unit: PERCENT. - [Average Performance Gap](): Lower is better; unit: SCALAR. - [Residual Training Error (RTE)](): Lower is better; unit: SCALAR. ## Recorded results | Method | Unlearning Accuracy (UA) | Result date | Report paper | | --- | --- | --- | --- | | RT | 100 | Not recorded | 68fa789e1e209131cea795fa | | RL | 100 | Not recorded | 68fa789e1e209131cea795fa | | SFRon | 100 | Not recorded | 68fa789e1e209131cea795fa | | SalUn | 100 | Not recorded | 68fa789e1e209131cea795fa | | MCUβ | 100 | Not recorded | 68fa789e1e209131cea795fa | | MCU | 99.96 | Not recorded | 68fa789e1e209131cea795fa | | NegGrad+ | 99.94 | Not recorded | 68fa789e1e209131cea795fa | | GA | 85.01 | Not recorded | 68fa789e1e209131cea795fa | | NegTV | 25.28 | Not recorded | 68fa789e1e209131cea795fa | | FT | 18.53 | Not recorded | 68fa789e1e209131cea795fa | ### RT Unlearning Accuracy (UA): 100. Result ID: 6a0450f647f34ce924eac9fb. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"PreResNet-110","Unlearning Setting":"Class-wise Forgetting"}. This record does not assert independent reproduction. ### RL Unlearning Accuracy (UA): 100. Result ID: 6a0450f647f34ce924eac9fd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"PreResNet-110","Unlearning Setting":"Class-wise Forgetting"}. This record does not assert independent reproduction. ### SFRon Unlearning Accuracy (UA): 100. Result ID: 6a0450f647f34ce924eaca00. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"PreResNet-110","Unlearning Setting":"Class-wise Forgetting"}. This record does not assert independent reproduction. ### SalUn Unlearning Accuracy (UA): 100. Result ID: 6a0450f647f34ce924eaca01. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"PreResNet-110","Unlearning Setting":"Class-wise Forgetting"}. This record does not assert independent reproduction. ### MCUβ Unlearning Accuracy (UA): 100. Result ID: 6a0450f647f34ce924eaca04. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"PreResNet-110","Unlearning Setting":"Class-wise Forgetting"}. This record does not assert independent reproduction. ### MCU Unlearning Accuracy (UA): 99.96. Result ID: 6a0450f647f34ce924eaca03. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"PreResNet-110","Unlearning Setting":"Class-wise Forgetting"}. This record does not assert independent reproduction. ### NegGrad+ Unlearning Accuracy (UA): 99.94. Result ID: 6a0450f647f34ce924eac9ff. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"PreResNet-110","Unlearning Setting":"Class-wise Forgetting"}. This record does not assert independent reproduction. ### GA Unlearning Accuracy (UA): 85.01. Result ID: 6a0450f647f34ce924eac9fe. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"PreResNet-110","Unlearning Setting":"Class-wise Forgetting"}. This record does not assert independent reproduction. ### NegTV Unlearning Accuracy (UA): 25.28. Result ID: 6a0450f647f34ce924eaca02. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"PreResNet-110","Unlearning Setting":"Class-wise Forgetting"}. This record does not assert independent reproduction. ### FT Unlearning Accuracy (UA): 18.53. Result ID: 6a0450f647f34ce924eac9fc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"PreResNet-110","Unlearning Setting":"Class-wise Forgetting"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.