# Knowledge Editing on UnKEBench Original questions Source: [Knowledge Editing on UnKEBench Original questions]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:36:42.491Z Evaluation of Knowledge Editing performance using the Original questions subset of the UnKEBench dataset. 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) | UnKEBench Original questions | | Metric | BERTScore | | Unit | PERCENT | | Direction | Higher 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 - [BERTScore](): Higher is better; unit: PERCENT. - [ROUGE-L](): Higher is better; unit: PERCENT. ## Recorded results | Method | BERTScore | Result date | Report paper | | --- | --- | --- | --- | | AnyEdit\* | 99.86 | Not recorded | 68fa7bfb1e209131cea8b9dd | | AnyEdit\* | 99.35 | Not recorded | 68fa7bfb1e209131cea8b9dd | | UnKE | 98.34 | Not recorded | 68fa7bfb1e209131cea8b9dd | | AnyEdit | 98.05 | Not recorded | 68fa7bfb1e209131cea8b9dd | | AnyEdit | 97.76 | Not recorded | 68fa7bfb1e209131cea8b9dd | | UnKE | 96.97 | Not recorded | 68fa7bfb1e209131cea8b9dd | | AlphaEdit | 80.48 | Not recorded | 68fa7bfb1e209131cea8b9dd | | MEMIT | 78.03 | Not recorded | 68fa7bfb1e209131cea8b9dd | | MEMIT | 76.21 | Not recorded | 68fa7bfb1e209131cea8b9dd | | ROME | 74.73 | Not recorded | 68fa7bfb1e209131cea8b9dd | | AlphaEdit | 73.92 | Not recorded | 68fa7bfb1e209131cea8b9dd | | ROME | 72.16 | Not recorded | 68fa7bfb1e209131cea8b9dd | | MEND | 69.49 | Not recorded | 68fa7bfb1e209131cea8b9dd | | MEND | 68.73 | Not recorded | 68fa7bfb1e209131cea8b9dd | | Pre-edited | 64.18 | Not recorded | 68fa7bfb1e209131cea8b9dd | | Pre-edited | 63.18 | Not recorded | 68fa7bfb1e209131cea8b9dd | | FT-L | 44.02 | Not recorded | 68fa7bfb1e209131cea8b9dd | | FT-L | 40.31 | Not recorded | 68fa7bfb1e209131cea8b9dd | ### AnyEdit\* BERTScore: 99.86. Result ID: 698b8d387a967eddb76f873c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"LLM":"Llama3-8B-It","Note":"Full layer gradient descent"}. This record does not assert independent reproduction. ### AnyEdit\* BERTScore: 99.35. Result ID: 698b8d387a967eddb76f8745. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"LLM":"Qwen2.5-7B-It","Note":"Full layer gradient descent"}. This record does not assert independent reproduction. ### UnKE BERTScore: 98.34. Result ID: 698b8d387a967eddb76f873b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM":"Llama3-8B-It"}. This record does not assert independent reproduction. ### AnyEdit BERTScore: 98.05. Result ID: 698b8d387a967eddb76f8743. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"LLM":"Qwen2.5-7B-It"}. This record does not assert independent reproduction. ### AnyEdit BERTScore: 97.76. Result ID: 698b8d387a967eddb76f873a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"LLM":"Llama3-8B-It"}. This record does not assert independent reproduction. ### UnKE BERTScore: 96.97. Result ID: 698b8d387a967eddb76f8744. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM":"Qwen2.5-7B-It"}. This record does not assert independent reproduction. ### AlphaEdit BERTScore: 80.48. Result ID: 698b8d387a967eddb76f8742. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM":"Qwen2.5-7B-It"}. This record does not assert independent reproduction. ### MEMIT BERTScore: 78.03. Result ID: 698b8d387a967eddb76f8741. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM":"Qwen2.5-7B-It"}. This record does not assert independent reproduction. ### MEMIT BERTScore: 76.21. Result ID: 698b8d387a967eddb76f8738. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM":"Llama3-8B-It"}. This record does not assert independent reproduction. ### ROME BERTScore: 74.73. Result ID: 698b8d387a967eddb76f8740. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"LLM":"Qwen2.5-7B-It"}. This record does not assert independent reproduction. ### AlphaEdit BERTScore: 73.92. Result ID: 698b8d387a967eddb76f8739. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM":"Llama3-8B-It"}. This record does not assert independent reproduction. ### ROME BERTScore: 72.16. Result ID: 698b8d387a967eddb76f8737. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"LLM":"Llama3-8B-It"}. This record does not assert independent reproduction. ### MEND BERTScore: 69.49. Result ID: 698b8d387a967eddb76f873f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM":"Qwen2.5-7B-It"}. This record does not assert independent reproduction. ### MEND BERTScore: 68.73. Result ID: 698b8d387a967eddb76f8736. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM":"Llama3-8B-It"}. This record does not assert independent reproduction. ### Pre-edited BERTScore: 64.18. Result ID: 698b8d387a967eddb76f873d. [Reporting paper]() Recorded evaluation configuration: {"LLM":"Qwen2.5-7B-It"}. This record does not assert independent reproduction. ### Pre-edited BERTScore: 63.18. Result ID: 698b8d387a967eddb76f8734. [Reporting paper]() Recorded evaluation configuration: {"LLM":"Llama3-8B-It"}. This record does not assert independent reproduction. ### FT-L BERTScore: 44.02. Result ID: 698b8d387a967eddb76f873e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM":"Qwen2.5-7B-It"}. This record does not assert independent reproduction. ### FT-L BERTScore: 40.31. Result ID: 698b8d387a967eddb76f8735. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM":"Llama3-8B-It"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.