# Mathematical Question Answering on GSM8K

Source: [Mathematical Question Answering on GSM8K](<https://www.sota2.com/research/sota/mathematical-question-answering-on-gsm8k>)

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

Data updated: 2026-03-26T14:25:31.489Z

Evaluation of model performance on the GSM8K dataset for mathematical question answering, measured using AUROC and 95% Confidence Intervals.

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 | Mathematical Question Answering |
| Dataset (including any stated split) | GSM8K |
| Metric | AUROC |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 3 of 3 |
| 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

- [AUROC](<https://www.sota2.com/research/sota/mathematical-question-answering-on-gsm8k.md?metric=AUROC>): Higher is better; unit: PERCENT.

- [AUROC 95% CI](<https://www.sota2.com/research/sota/mathematical-question-answering-on-gsm8k.md?metric=AUROC%2095%25%20CI>): Higher is better; unit: PERCENT.

## Recorded results

| Method | AUROC | Result date | Report paper |
| --- | --- | --- | --- |
| B1 max entropy | 72.4 | Not recorded | 69c4bdb60636e5bf4afc7760 |
| Combined (4 entropy feat, CV) | 72.4 | Not recorded | 69c4bdb60636e5bf4afc7760 |
| P(True) | 60.8 | Not recorded | 69c4bdb60636e5bf4afc7760 |

### B1 max entropy

AUROC: 72.4. Result ID: 69c541bf47f34ce924af22f4.

[Reporting paper](<https://www.sota2.com/research/paper/69c4bdb60636e5bf4afc7760.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c4bdb60636e5bf4afc7760.md>)

Recorded evaluation configuration: {"Cost":"Free"}. This record does not assert independent reproduction.

### Combined (4 entropy feat, CV)

AUROC: 72.4. Result ID: 69c541bf47f34ce924af22f5.

[Reporting paper](<https://www.sota2.com/research/paper/69c4bdb60636e5bf4afc7760.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c4bdb60636e5bf4afc7760.md>)

Recorded evaluation configuration: {"Cost":"Free"}. This record does not assert independent reproduction.

### P(True)

AUROC: 60.8. Result ID: 69c541bf47f34ce924af22f6.

[Reporting paper](<https://www.sota2.com/research/paper/69c4bdb60636e5bf4afc7760.md>)

Recorded evaluation configuration: {"Cost":"1 call"}. This record does not assert independent reproduction.

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