# Fact Extraction and Verification on FEVER (dev)

Source: [Fact Extraction and Verification on FEVER (dev)](<https://www.sota2.com/research/sota/fact-extraction-and-verification-on-fever-dev>)

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

Data updated: 2026-02-26T05:48:24.028Z

Evaluation of model performance on the development split of the FEVER dataset for Fact Extraction and Verification, reporting F1, LA, and FS 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 | Fact Extraction and Verification |
| Dataset (including any stated split) | FEVER (dev) |
| Metric | Label Accuracy (LA) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 9 of 9 |
| 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

- [Label Accuracy (LA)](<https://www.sota2.com/research/sota/fact-extraction-and-verification-on-fever-dev.md?metric=Label%20Accuracy%20(LA)>): Higher is better; unit: PERCENT.

- [F1 Score](<https://www.sota2.com/research/sota/fact-extraction-and-verification-on-fever-dev.md?metric=F1%20Score>): Higher is better; unit: PERCENT.

- [Fact Score (FS)](<https://www.sota2.com/research/sota/fact-extraction-and-verification-on-fever-dev.md?metric=Fact%20Score%20(FS)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Label Accuracy (LA) | Result date | Report paper |
| --- | --- | --- | --- |
| CICD | 76.3 | Not recorded | 68fa6a051e209131cea257af |
| Full system (single) | 75.12 | Not recorded | 68fa66641e209131cea100b8 |
| KGAT | 74.5 | Not recorded | 68fa6a051e209131cea257af |
| GEAR | 73.8 | Not recorded | 68fa6a051e209131cea257af |
| HAN | 72 | Not recorded | 68fa6a051e209131cea257af |
| NSMN | 69.7 | Not recorded | 68fa6a051e209131cea257af |
| Yoneda (2018) | 69.66 | Not recorded | 68fa66641e209131cea100b8 |
| Nie (2019) | 69.64 | Not recorded | 68fa66641e209131cea100b8 |
| Hanselowski (2018) | 68.49 | Not recorded | 68fa66641e209131cea100b8 |

### CICD

Label Accuracy (LA): 76.3. Result ID: 698dfec4ffc3901de09a0504.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a051e209131cea257af.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a051e209131cea257af.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Full system (single)

Label Accuracy (LA): 75.12. Result ID: 698cef3b8b517eb4331dce94.

[Reporting paper](<https://www.sota2.com/research/paper/68fa66641e209131cea100b8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66641e209131cea100b8.md>)

[Implementation](<https://github.com/easonnie/semanticRetrievalMRS>)

Recorded evaluation configuration: {"Mode":"single"}. This record does not assert independent reproduction.

### KGAT

Label Accuracy (LA): 74.5. Result ID: 698dfec4ffc3901de09a0503.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a051e209131cea257af.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a3c1e209131cea26da8.md>)

[Implementation](<https://github.com/thunlp/KernelGAT>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### GEAR

Label Accuracy (LA): 73.8. Result ID: 698dfec4ffc3901de09a0502.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a051e209131cea257af.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa661b1e209131cea0e57c.md>)

[Implementation](<https://github.com/thunlp/GEAR>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### HAN

Label Accuracy (LA): 72. Result ID: 698dfec4ffc3901de09a0501.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a051e209131cea257af.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### NSMN

Label Accuracy (LA): 69.7. Result ID: 698dfec4ffc3901de09a0500.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a051e209131cea257af.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65121e209131cea07fe3.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Yoneda (2018)

Label Accuracy (LA): 69.66. Result ID: 698cef3b8b517eb4331dce92.

[Reporting paper](<https://www.sota2.com/research/paper/68fa66641e209131cea100b8.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Nie (2019)

Label Accuracy (LA): 69.64. Result ID: 698cef3b8b517eb4331dce93.

[Reporting paper](<https://www.sota2.com/research/paper/68fa66641e209131cea100b8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65121e209131cea07fe3.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Hanselowski (2018)

Label Accuracy (LA): 68.49. Result ID: 698cef3b8b517eb4331dce91.

[Reporting paper](<https://www.sota2.com/research/paper/68fa66641e209131cea100b8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65c11e209131cea0c3ad.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

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