# Knowledge Editing on Counterfact 10,000 facts

Source: [Knowledge Editing on Counterfact 10,000 facts](<https://www.sota2.com/research/sota/knowledge-editing-on-counterfact-10-000-facts>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T06:01:52.253Z

Evaluation of Knowledge Editing performance on the Counterfact 10,000 facts dataset, measured by Relational (Rel.), Generality (Gen.), and Locality (Loc.) scores.

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) | Counterfact 10,000 facts |
| Metric | Relational Score |
| 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

- [Relational Score](<https://www.sota2.com/research/sota/knowledge-editing-on-counterfact-10-000-facts.md?metric=Relational%20Score>): Higher is better; unit: PERCENT.

- [Generality Score](<https://www.sota2.com/research/sota/knowledge-editing-on-counterfact-10-000-facts.md?metric=Generality%20Score>): Higher is better; unit: PERCENT.

- [Locality Score](<https://www.sota2.com/research/sota/knowledge-editing-on-counterfact-10-000-facts.md?metric=Locality%20Score>): Higher is better; unit: PERCENT.

- [Average Score](<https://www.sota2.com/research/sota/knowledge-editing-on-counterfact-10-000-facts.md?metric=Average%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Relational Score | Result date | Report paper |
| --- | --- | --- | --- |
| GRACE | 10000 | Not recorded | 68fa77481e209131cea72408 |
| GRACE | 10000 | Not recorded | 68fa77481e209131cea72408 |
| GRACE | 10000 | Not recorded | 68fa77481e209131cea72408 |
| MEMIT | 9580 | Not recorded | 68fa77481e209131cea72408 |
| MindBridge | 9560 | Not recorded | 68fa77481e209131cea72408 |
| MindBridge | 9460 | Not recorded | 68fa77481e209131cea72408 |
| MindBridge | 9385 | Not recorded | 68fa77481e209131cea72408 |
| AlphaEdit | 9200 | Not recorded | 68fa77481e209131cea72408 |
| AlphaEdit | 9170 | Not recorded | 68fa77481e209131cea72408 |
| EMMET | 8551 | Not recorded | 68fa77481e209131cea72408 |
| EMMET | 8551 | Not recorded | 68fa77481e209131cea72408 |
| MEMIT | 8471 | Not recorded | 68fa77481e209131cea72408 |
| AlphaEdit | 6600 | Not recorded | 68fa77481e209131cea72408 |
| EMMET | 6053 | Not recorded | 68fa77481e209131cea72408 |
| MEMIT | 4536 | Not recorded | 68fa77481e209131cea72408 |
| FT-L | 2567 | Not recorded | 68fa77481e209131cea72408 |
| WISE | 1838 | Not recorded | 68fa77481e209131cea72408 |
| WISE | 1608 | Not recorded | 68fa77481e209131cea72408 |
| FT-L | 1058 | Not recorded | 68fa77481e209131cea72408 |
| WISE | 179 | Not recorded | 68fa77481e209131cea72408 |
| Pre-edited | 70 | Not recorded | 68fa77481e209131cea72408 |
| Pre-edited | 28 | Not recorded | 68fa77481e209131cea72408 |
| Pre-edited | 9 | Not recorded | 68fa77481e209131cea72408 |
| r-ROME | 9 | Not recorded | 68fa77481e209131cea72408 |
| FT-L | 0 | Not recorded | 68fa77481e209131cea72408 |
| r-ROME | 0 | Not recorded | 68fa77481e209131cea72408 |
| r-ROME | 0 | Not recorded | 68fa77481e209131cea72408 |

### GRACE

Relational Score: 10000. Result ID: 69948f663a6c3148491e06e5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77481e209131cea72408.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70ef1e209131cea4e231.md>)

[Implementation](<https://www.sota2.com/github.com/thartvigsen/grace>)

Recorded evaluation configuration: {"Model":"GPT-XL"}. This record does not assert independent reproduction.

### GRACE

Relational Score: 10000. Result ID: 69948f663a6c3148491e06ee.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77481e209131cea72408.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70ef1e209131cea4e231.md>)

[Implementation](<https://www.sota2.com/github.com/thartvigsen/grace>)

Recorded evaluation configuration: {"Model":"GPT-J"}. This record does not assert independent reproduction.

### GRACE

Relational Score: 10000. Result ID: 69948f663a6c3148491e06f7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77481e209131cea72408.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70ef1e209131cea4e231.md>)

[Implementation](<https://www.sota2.com/github.com/thartvigsen/grace>)

Recorded evaluation configuration: {"Model":"LLaMA3"}. This record does not assert independent reproduction.

### MEMIT

Relational Score: 9580. Result ID: 69948f663a6c3148491e06f0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77481e209131cea72408.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa703d1e209131cea49dc7.md>)

Recorded evaluation configuration: {"Model":"GPT-J"}. This record does not assert independent reproduction.

### MindBridge

Relational Score: 9560. Result ID: 69948f663a6c3148491e06e8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77481e209131cea72408.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77481e209131cea72408.md>)

[Implementation](<https://github.com/CrashBugger/MindBridge>)

Recorded evaluation configuration: {"Model":"GPT-XL"}. This record does not assert independent reproduction.

### MindBridge

Relational Score: 9460. Result ID: 69948f663a6c3148491e06f1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77481e209131cea72408.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77481e209131cea72408.md>)

[Implementation](<https://github.com/CrashBugger/MindBridge>)

Recorded evaluation configuration: {"Model":"GPT-J"}. This record does not assert independent reproduction.

### MindBridge

Relational Score: 9385. Result ID: 69948f663a6c3148491e06fa.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77481e209131cea72408.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77481e209131cea72408.md>)

[Implementation](<https://github.com/CrashBugger/MindBridge>)

Recorded evaluation configuration: {"Model":"LLaMA3"}. This record does not assert independent reproduction.

### AlphaEdit

Relational Score: 9200. Result ID: 69948f663a6c3148491e06ec.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77481e209131cea72408.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78351e209131cea77370.md>)

Recorded evaluation configuration: {"Model":"GPT-J"}. This record does not assert independent reproduction.

### AlphaEdit

Relational Score: 9170. Result ID: 69948f663a6c3148491e06f5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77481e209131cea72408.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78351e209131cea77370.md>)

Recorded evaluation configuration: {"Model":"LLaMA3"}. This record does not assert independent reproduction.

### EMMET

Relational Score: 8551. Result ID: 69948f663a6c3148491e06e4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77481e209131cea72408.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75221e209131cea660a0.md>)

Recorded evaluation configuration: {"Model":"GPT-XL"}. This record does not assert independent reproduction.

### EMMET

Relational Score: 8551. Result ID: 69948f663a6c3148491e06ed.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77481e209131cea72408.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75221e209131cea660a0.md>)

Recorded evaluation configuration: {"Model":"GPT-J"}. This record does not assert independent reproduction.

### MEMIT

Relational Score: 8471. Result ID: 69948f663a6c3148491e06f9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77481e209131cea72408.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa703d1e209131cea49dc7.md>)

Recorded evaluation configuration: {"Model":"LLaMA3"}. This record does not assert independent reproduction.

### AlphaEdit

Relational Score: 6600. Result ID: 69948f663a6c3148491e06e3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77481e209131cea72408.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78351e209131cea77370.md>)

Recorded evaluation configuration: {"Model":"GPT-XL"}. This record does not assert independent reproduction.

### EMMET

Relational Score: 6053. Result ID: 69948f663a6c3148491e06f6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77481e209131cea72408.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75221e209131cea660a0.md>)

Recorded evaluation configuration: {"Model":"LLaMA3"}. This record does not assert independent reproduction.

### MEMIT

Relational Score: 4536. Result ID: 69948f663a6c3148491e06e7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77481e209131cea72408.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa703d1e209131cea49dc7.md>)

Recorded evaluation configuration: {"Model":"GPT-XL"}. This record does not assert independent reproduction.

### FT-L

Relational Score: 2567. Result ID: 69948f663a6c3148491e06f3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77481e209131cea72408.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e591e209131cea3f3c9.md>)

[Implementation](<https://rome.baulab.info/>)

Recorded evaluation configuration: {"Model":"LLaMA3"}. This record does not assert independent reproduction.

### WISE

Relational Score: 1838. Result ID: 69948f663a6c3148491e06eb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77481e209131cea72408.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76461e209131cea6c143.md>)

Recorded evaluation configuration: {"Model":"GPT-J"}. This record does not assert independent reproduction.

### WISE

Relational Score: 1608. Result ID: 69948f663a6c3148491e06f4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77481e209131cea72408.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76461e209131cea6c143.md>)

Recorded evaluation configuration: {"Model":"LLaMA3"}. This record does not assert independent reproduction.

### FT-L

Relational Score: 1058. Result ID: 69948f663a6c3148491e06ea.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77481e209131cea72408.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e591e209131cea3f3c9.md>)

[Implementation](<https://rome.baulab.info/>)

Recorded evaluation configuration: {"Model":"GPT-J"}. This record does not assert independent reproduction.

### WISE

Relational Score: 179. Result ID: 69948f663a6c3148491e06e2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77481e209131cea72408.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76461e209131cea6c143.md>)

Recorded evaluation configuration: {"Model":"GPT-XL"}. This record does not assert independent reproduction.

### Pre-edited

Relational Score: 70. Result ID: 69948f663a6c3148491e06f2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77481e209131cea72408.md>)

Recorded evaluation configuration: {"Model":"LLaMA3"}. This record does not assert independent reproduction.

### Pre-edited

Relational Score: 28. Result ID: 69948f663a6c3148491e06e0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77481e209131cea72408.md>)

Recorded evaluation configuration: {"Model":"GPT-XL"}. This record does not assert independent reproduction.

### Pre-edited

Relational Score: 9. Result ID: 69948f663a6c3148491e06e9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77481e209131cea72408.md>)

Recorded evaluation configuration: {"Model":"GPT-J"}. This record does not assert independent reproduction.

### r-ROME

Relational Score: 9. Result ID: 69948f663a6c3148491e06f8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77481e209131cea72408.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75211e209131cea66050.md>)

Recorded evaluation configuration: {"Model":"LLaMA3"}. This record does not assert independent reproduction.

### FT-L

Relational Score: 0. Result ID: 69948f663a6c3148491e06e1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77481e209131cea72408.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e591e209131cea3f3c9.md>)

[Implementation](<https://rome.baulab.info/>)

Recorded evaluation configuration: {"Model":"GPT-XL"}. This record does not assert independent reproduction.

### r-ROME

Relational Score: 0. Result ID: 69948f663a6c3148491e06e6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77481e209131cea72408.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75211e209131cea66050.md>)

Recorded evaluation configuration: {"Model":"GPT-XL"}. This record does not assert independent reproduction.

### r-ROME

Relational Score: 0. Result ID: 69948f663a6c3148491e06ef.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77481e209131cea72408.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75211e209131cea66050.md>)

Recorded evaluation configuration: {"Model":"GPT-J"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.