# Knowledge Unlearning on FaithUn Source: [Knowledge Unlearning on FaithUn]() Collection index: [llms.txt]() Data updated: 2026-03-05T06:57:18.216Z Evaluation of Knowledge Unlearning performance on the FaithUn dataset, reporting metrics related to forgetting rates (FS, RS, US) and attack robustness. 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 | Knowledge Unlearning | | Dataset (including any stated split) | FaithUn | | Metric | FS | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 16 of 16 | | 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 - [FS](): Higher is better; unit: PERCENT. - [RS](): Higher is better; unit: PERCENT. - [US](): Higher is better; unit: PERCENT. - [Harmful Attack Success Rate (p=0.1)](): Lower is better; unit: PERCENT. - [Harmful Attack Success Rate (p=0.3)](): Lower is better; unit: PERCENT. - [Benign Attack Impact](): Lower is better; unit: PERCENT. ## Recorded results | Method | FS | Result date | Report paper | | --- | --- | --- | --- | | Default | 91.92 | Not recorded | 68fa7baa1e209131cea89c3d | | Default | 78.79 | Not recorded | 68fa7baa1e209131cea89c3d | | GA | 0 | Not recorded | 68fa7baa1e209131cea89c3d | | GD | 0 | Not recorded | 68fa7baa1e209131cea89c3d | | DPO | 0 | Not recorded | 68fa7baa1e209131cea89c3d | | NPO | 0 | Not recorded | 68fa7baa1e209131cea89c3d | | RMU | 0 | Not recorded | 68fa7baa1e209131cea89c3d | | KLUE | 0 | Not recorded | 68fa7baa1e209131cea89c3d | | SSIUU | 0 | Not recorded | 68fa7baa1e209131cea89c3d | | GA | 0 | Not recorded | 68fa7baa1e209131cea89c3d | | GD | 0 | Not recorded | 68fa7baa1e209131cea89c3d | | DPO | 0 | Not recorded | 68fa7baa1e209131cea89c3d | | NPO | 0 | Not recorded | 68fa7baa1e209131cea89c3d | | RMU | 0 | Not recorded | 68fa7baa1e209131cea89c3d | | KLUE | 0 | Not recorded | 68fa7baa1e209131cea89c3d | | SSIUU | 0 | Not recorded | 68fa7baa1e209131cea89c3d | ### Default FS: 91.92. Result ID: 69a927df6dbb7075b951f47e. [Reporting paper]() Recorded evaluation configuration: {"Model":"Llama-3.2"}. This record does not assert independent reproduction. ### Default FS: 78.79. Result ID: 69a927df6dbb7075b951f486. [Reporting paper]() Recorded evaluation configuration: {"Model":"Qwen-2.5"}. This record does not assert independent reproduction. ### GA FS: 0. Result ID: 69a927df6dbb7075b951f47f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"Llama-3.2"}. This record does not assert independent reproduction. ### GD FS: 0. Result ID: 69a927df6dbb7075b951f480. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama-3.2"}. This record does not assert independent reproduction. ### DPO FS: 0. Result ID: 69a927df6dbb7075b951f481. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama-3.2"}. This record does not assert independent reproduction. ### NPO FS: 0. Result ID: 69a927df6dbb7075b951f482. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama-3.2"}. This record does not assert independent reproduction. ### RMU FS: 0. Result ID: 69a927df6dbb7075b951f483. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama-3.2"}. This record does not assert independent reproduction. ### KLUE FS: 0. Result ID: 69a927df6dbb7075b951f484. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama-3.2"}. This record does not assert independent reproduction. ### SSIUU FS: 0. Result ID: 69a927df6dbb7075b951f485. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama-3.2"}. This record does not assert independent reproduction. ### GA FS: 0. Result ID: 69a927df6dbb7075b951f487. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"Qwen-2.5"}. This record does not assert independent reproduction. ### GD FS: 0. Result ID: 69a927df6dbb7075b951f488. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen-2.5"}. This record does not assert independent reproduction. ### DPO FS: 0. Result ID: 69a927df6dbb7075b951f489. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen-2.5"}. This record does not assert independent reproduction. ### NPO FS: 0. Result ID: 69a927df6dbb7075b951f48a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen-2.5"}. This record does not assert independent reproduction. ### RMU FS: 0. Result ID: 69a927df6dbb7075b951f48b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen-2.5"}. This record does not assert independent reproduction. ### KLUE FS: 0. Result ID: 69a927df6dbb7075b951f48c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen-2.5"}. This record does not assert independent reproduction. ### SSIUU FS: 0. Result ID: 69a927df6dbb7075b951f48d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen-2.5"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.