# Fact Verification on FEVER (Accuracy, Selectivity, and Abstention)

Source: [Fact Verification on FEVER (Accuracy, Selectivity, and Abstention)](<https://www.sota2.com/research/sota/fact-verification-on-fever-accuracy-selectivity-and-abstention>)

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

Data updated: 2026-02-26T05:33:07.142Z

Evaluation of model performance on the FEVER dataset for Fact Verification, focusing on raw accuracy alongside metrics related to selective prediction and model abstention behavior.

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 Verification |
| Dataset (including any stated split) | FEVER |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 2 of 2 |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/fact-verification-on-fever-accuracy-selectivity-and-abstention.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [Selectivity (α=0.2)](<https://www.sota2.com/research/sota/fact-verification-on-fever-accuracy-selectivity-and-abstention.md?metric=Selectivity%20(%CE%B1%3D0.2)>): Higher is better; unit: SCALAR.

- [Abstention Rate](<https://www.sota2.com/research/sota/fact-verification-on-fever-accuracy-selectivity-and-abstention.md?metric=Abstention%20Rate>): Lower is better; unit: PERCENT.

- [Abstention Precision](<https://www.sota2.com/research/sota/fact-verification-on-fever-accuracy-selectivity-and-abstention.md?metric=Abstention%20Precision>): Higher is better; unit: PERCENT.

- [Stability (Std Dev)](<https://www.sota2.com/research/sota/fact-verification-on-fever-accuracy-selectivity-and-abstention.md?metric=Stability%20(Std%20Dev)>): Lower is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| MMA | 59.93 | Not recorded | 6996990909c33900fcd71efc |
| MIRIX | 59.87 | Not recorded | 6996990909c33900fcd71efc |

### MMA

Accuracy: 59.93. Result ID: 699af1965698665ec482b620.

[Reporting paper](<https://www.sota2.com/research/paper/6996990909c33900fcd71efc.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6996990909c33900fcd71efc.md>)

[Implementation](<https://github.com/AIGeeksGroup/MMA>)

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

### MIRIX

Accuracy: 59.87. Result ID: 699af1965698665ec482b61f.

[Reporting paper](<https://www.sota2.com/research/paper/6996990909c33900fcd71efc.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79fb1e209131cea80723.md>)

Recorded evaluation configuration: {"Baseline":"true"}. This record does not assert independent reproduction.

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