# Knowledge Editing on ZsRE 10,000 facts Source: [Knowledge Editing on ZsRE 10,000 facts]() Collection index: [llms.txt]() Data updated: 2026-02-26T06:02:05.299Z Evaluation of Knowledge Editing performance on the ZsRE dataset using the specific '10,000 facts' configuration, measured using Reliability (Rel.), Generality (Gen.), and Locality (Loc.) metrics. 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 Editing | | Dataset (including any stated split) | ZsRE 10,000 facts | | Metric | Reliability | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 27 of 27 | | 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 - [Reliability](): Higher is better; unit: PERCENT. - [Generality](): Higher is better; unit: PERCENT. - [Locality](): Higher is better; unit: PERCENT. - [Average Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | Reliability | Result date | Report paper | | --- | --- | --- | --- | | GRACE | 100 | Not recorded | 68fa77481e209131cea72408 | | GRACE | 100 | Not recorded | 68fa77481e209131cea72408 | | GRACE | 100 | Not recorded | 68fa77481e209131cea72408 | | MindBridge | 99.06 | Not recorded | 68fa77481e209131cea72408 | | MindBridge | 99.03 | Not recorded | 68fa77481e209131cea72408 | | MEMIT | 86.17 | Not recorded | 68fa77481e209131cea72408 | | AlphaEdit | 78.64 | Not recorded | 68fa77481e209131cea72408 | | MindBridge | 78.14 | Not recorded | 68fa77481e209131cea72408 | | AlphaEdit | 76.95 | Not recorded | 68fa77481e209131cea72408 | | EMMET | 70.37 | Not recorded | 68fa77481e209131cea72408 | | EMMET | 70.37 | Not recorded | 68fa77481e209131cea72408 | | MEMIT | 66.16 | Not recorded | 68fa77481e209131cea72408 | | EMMET | 62.94 | Not recorded | 68fa77481e209131cea72408 | | AlphaEdit | 51.82 | Not recorded | 68fa77481e209131cea72408 | | MEMIT | 47.2 | Not recorded | 68fa77481e209131cea72408 | | WISE | 36.88 | Not recorded | 68fa77481e209131cea72408 | | WISE | 29.18 | Not recorded | 68fa77481e209131cea72408 | | Pre-edited | 27.7 | Not recorded | 68fa77481e209131cea72408 | | WISE | 24.88 | Not recorded | 68fa77481e209131cea72408 | | Pre-edited | 23.01 | Not recorded | 68fa77481e209131cea72408 | | Pre-edited | 22.44 | Not recorded | 68fa77481e209131cea72408 | | FT-L | 15.73 | Not recorded | 68fa77481e209131cea72408 | | FT-L | 15.6 | Not recorded | 68fa77481e209131cea72408 | | FT-L | 5.99 | Not recorded | 68fa77481e209131cea72408 | | r-ROME | 0.7 | Not recorded | 68fa77481e209131cea72408 | | r-ROME | 0.09 | Not recorded | 68fa77481e209131cea72408 | | r-ROME | 0 | Not recorded | 68fa77481e209131cea72408 | ### GRACE Reliability: 100. Result ID: 69948f663a6c3148491e0700. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"GPT-XL"}. This record does not assert independent reproduction. ### GRACE Reliability: 100. Result ID: 69948f663a6c3148491e0709. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"GPT-J"}. This record does not assert independent reproduction. ### GRACE Reliability: 100. Result ID: 69948f663a6c3148491e0712. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"LLaMA3"}. This record does not assert independent reproduction. ### MindBridge Reliability: 99.06. Result ID: 69948f663a6c3148491e070c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"GPT-J"}. This record does not assert independent reproduction. ### MindBridge Reliability: 99.03. Result ID: 69948f663a6c3148491e0715. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"LLaMA3"}. This record does not assert independent reproduction. ### MEMIT Reliability: 86.17. Result ID: 69948f663a6c3148491e070b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-J"}. This record does not assert independent reproduction. ### AlphaEdit Reliability: 78.64. Result ID: 69948f663a6c3148491e0710. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaMA3"}. This record does not assert independent reproduction. ### MindBridge Reliability: 78.14. Result ID: 69948f663a6c3148491e0703. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"GPT-XL"}. This record does not assert independent reproduction. ### AlphaEdit Reliability: 76.95. Result ID: 69948f663a6c3148491e0707. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-J"}. This record does not assert independent reproduction. ### EMMET Reliability: 70.37. Result ID: 69948f663a6c3148491e06ff. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-XL"}. This record does not assert independent reproduction. ### EMMET Reliability: 70.37. Result ID: 69948f663a6c3148491e0708. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-J"}. This record does not assert independent reproduction. ### MEMIT Reliability: 66.16. Result ID: 69948f663a6c3148491e0714. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaMA3"}. This record does not assert independent reproduction. ### EMMET Reliability: 62.94. Result ID: 69948f663a6c3148491e0711. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaMA3"}. This record does not assert independent reproduction. ### AlphaEdit Reliability: 51.82. Result ID: 69948f663a6c3148491e06fe. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-XL"}. This record does not assert independent reproduction. ### MEMIT Reliability: 47.2. Result ID: 69948f663a6c3148491e0702. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-XL"}. This record does not assert independent reproduction. ### WISE Reliability: 36.88. Result ID: 69948f663a6c3148491e0706. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-J"}. This record does not assert independent reproduction. ### WISE Reliability: 29.18. Result ID: 69948f663a6c3148491e070f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaMA3"}. This record does not assert independent reproduction. ### Pre-edited Reliability: 27.7. Result ID: 69948f663a6c3148491e070d. [Reporting paper]() Recorded evaluation configuration: {"Model":"LLaMA3"}. This record does not assert independent reproduction. ### WISE Reliability: 24.88. Result ID: 69948f663a6c3148491e06fd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-XL"}. This record does not assert independent reproduction. ### Pre-edited Reliability: 23.01. Result ID: 69948f663a6c3148491e0704. [Reporting paper]() Recorded evaluation configuration: {"Model":"GPT-J"}. This record does not assert independent reproduction. ### Pre-edited Reliability: 22.44. Result ID: 69948f663a6c3148491e06fb. [Reporting paper]() Recorded evaluation configuration: {"Model":"GPT-XL"}. This record does not assert independent reproduction. ### FT-L Reliability: 15.73. Result ID: 69948f663a6c3148491e0705. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"GPT-J"}. This record does not assert independent reproduction. ### FT-L Reliability: 15.6. Result ID: 69948f663a6c3148491e06fc. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"GPT-XL"}. This record does not assert independent reproduction. ### FT-L Reliability: 5.99. Result ID: 69948f663a6c3148491e070e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"LLaMA3"}. This record does not assert independent reproduction. ### r-ROME Reliability: 0.7. Result ID: 69948f663a6c3148491e0713. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaMA3"}. This record does not assert independent reproduction. ### r-ROME Reliability: 0.09. Result ID: 69948f663a6c3148491e070a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-J"}. This record does not assert independent reproduction. ### r-ROME Reliability: 0. Result ID: 69948f663a6c3148491e0701. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-XL"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.