# Machine Unlearning on TOFU 1% (forget set) Source: [Machine Unlearning on TOFU 1% (forget set)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:44:59.171Z Evaluation of Machine Unlearning effectiveness on the 1% subset of the TOFU dataset, specifically focusing on results derived from the Forget Set configuration. 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) | TOFU 1% (forget set) | | Metric | FQ | | Unit | SCALAR | | Direction | Lower is better | | Results loaded | 18 of 18 | | 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 - [FQ](): Lower is better; unit: SCALAR. - [MU Score](): Higher is better; unit: PERCENT. - [ES (Df)](): Higher is better; unit: PERCENT. - [ES (Dr)](): Higher is better; unit: PERCENT. - [Gib Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | FQ | Result date | Report paper | | --- | --- | --- | --- | | Full | -2.17 | Not recorded | 697c3b76a8b01e76cc8da7b2 | | NPO | -2.062 | Not recorded | 697c3b76a8b01e76cc8da7b2 | | GA | -1.953 | Not recorded | 697c3b76a8b01e76cc8da7b2 | | GD | -1.845 | Not recorded | 697c3b76a8b01e76cc8da7b2 | | NPO+ | -1.845 | Not recorded | 697c3b76a8b01e76cc8da7b2 | | Full | -1.845 | Not recorded | 697c3b76a8b01e76cc8da7b2 | | GA | -1.845 | Not recorded | 697c3b76a8b01e76cc8da7b2 | | GD | -1.845 | Not recorded | 697c3b76a8b01e76cc8da7b2 | | NPO | -1.845 | Not recorded | 697c3b76a8b01e76cc8da7b2 | | NPO+ | -1.845 | Not recorded | 697c3b76a8b01e76cc8da7b2 | | PerTA-fisher | -0.576 | Not recorded | 697c3b76a8b01e76cc8da7b2 | | TV | -0.393 | Not recorded | 697c3b76a8b01e76cc8da7b2 | | PerTA-grad | -0.289 | Not recorded | 697c3b76a8b01e76cc8da7b2 | | TV | -0.238 | Not recorded | 697c3b76a8b01e76cc8da7b2 | | PerTA-fisher | -0.238 | Not recorded | 697c3b76a8b01e76cc8da7b2 | | PerTA-grad | -0.037 | Not recorded | 697c3b76a8b01e76cc8da7b2 | | GT | 0 | Not recorded | 697c3b76a8b01e76cc8da7b2 | | GT | 0 | Not recorded | 697c3b76a8b01e76cc8da7b2 | ### Full FQ: -2.17. Result ID: 6998f6125698665ec45b1bc0. [Reporting paper]() Recorded evaluation configuration: {"Model Size":"1B","Method Category":"Reference"}. This record does not assert independent reproduction. ### NPO FQ: -2.062. Result ID: 6998f6125698665ec45b1bc4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"1B","Method Category":"Training-based"}. This record does not assert independent reproduction. ### GA FQ: -1.953. Result ID: 6998f6125698665ec45b1bc2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"1B","Method Category":"Training-based"}. This record does not assert independent reproduction. ### GD FQ: -1.845. Result ID: 6998f6125698665ec45b1bc3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"1B","Method Category":"Training-based"}. This record does not assert independent reproduction. ### NPO+ FQ: -1.845. Result ID: 6998f6125698665ec45b1bc5. [Reporting paper]() Recorded evaluation configuration: {"Model Size":"1B","Method Category":"Training-based"}. This record does not assert independent reproduction. ### Full FQ: -1.845. Result ID: 6998f6125698665ec45b1bc9. [Reporting paper]() Recorded evaluation configuration: {"Model Size":"3B","Method Category":"Reference"}. This record does not assert independent reproduction. ### GA FQ: -1.845. Result ID: 6998f6125698665ec45b1bcb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"3B","Method Category":"Training-based"}. This record does not assert independent reproduction. ### GD FQ: -1.845. Result ID: 6998f6125698665ec45b1bcc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"3B","Method Category":"Training-based"}. This record does not assert independent reproduction. ### NPO FQ: -1.845. Result ID: 6998f6125698665ec45b1bcd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"3B","Method Category":"Training-based"}. This record does not assert independent reproduction. ### NPO+ FQ: -1.845. Result ID: 6998f6125698665ec45b1bce. [Reporting paper]() Recorded evaluation configuration: {"Model Size":"3B","Method Category":"Training-based"}. This record does not assert independent reproduction. ### PerTA-fisher FQ: -0.576. Result ID: 6998f6125698665ec45b1bc8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"1B","Method Category":"Task Arithmetic-based"}. This record does not assert independent reproduction. ### TV FQ: -0.393. Result ID: 6998f6125698665ec45b1bc6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"1B","Method Category":"Task Arithmetic-based"}. This record does not assert independent reproduction. ### PerTA-grad FQ: -0.289. Result ID: 6998f6125698665ec45b1bc7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"1B","Method Category":"Task Arithmetic-based"}. This record does not assert independent reproduction. ### TV FQ: -0.238. Result ID: 6998f6125698665ec45b1bcf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"3B","Method Category":"Task Arithmetic-based"}. This record does not assert independent reproduction. ### PerTA-fisher FQ: -0.238. Result ID: 6998f6125698665ec45b1bd1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"3B","Method Category":"Task Arithmetic-based"}. This record does not assert independent reproduction. ### PerTA-grad FQ: -0.037. Result ID: 6998f6125698665ec45b1bd0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"3B","Method Category":"Task Arithmetic-based"}. This record does not assert independent reproduction. ### GT FQ: 0. Result ID: 6998f6125698665ec45b1bc1. [Reporting paper]() Recorded evaluation configuration: {"Model Size":"1B","Method Category":"Reference"}. This record does not assert independent reproduction. ### GT FQ: 0. Result ID: 6998f6125698665ec45b1bca. [Reporting paper]() Recorded evaluation configuration: {"Model Size":"3B","Method Category":"Reference"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.