# Mathematical Reasoning on MATH (test) (Sub-category Scores)

Source: [Mathematical Reasoning on MATH (test) (Sub-category Scores)](<https://www.sota2.com/research/sota/mathematical-reasoning-on-math-test-sub-category-scores>)

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

Data updated: 2026-05-27T09:32:05.105Z

Evaluation of model performance on the test split of the MATH dataset, reporting accuracy scores across specific mathematical domains (Algebra, Geometry, Number Theory, etc.).

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 Reasoning |
| Dataset (including any stated split) | MATH (test) |
| Metric | Algebra Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 17 of 17 |
| 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

- [Algebra Accuracy](<https://www.sota2.com/research/sota/mathematical-reasoning-on-math-test-sub-category-scores.md?metric=Algebra%20Accuracy>): Higher is better; unit: PERCENT.

- [Counting & Probability Accuracy](<https://www.sota2.com/research/sota/mathematical-reasoning-on-math-test-sub-category-scores.md?metric=Counting%20%26%20Probability%20Accuracy>): Higher is better; unit: PERCENT.

- [Geometry Accuracy](<https://www.sota2.com/research/sota/mathematical-reasoning-on-math-test-sub-category-scores.md?metric=Geometry%20Accuracy>): Higher is better; unit: PERCENT.

- [Intermediate Algebra Accuracy](<https://www.sota2.com/research/sota/mathematical-reasoning-on-math-test-sub-category-scores.md?metric=Intermediate%20Algebra%20Accuracy>): Higher is better; unit: PERCENT.

- [Number Theory Accuracy](<https://www.sota2.com/research/sota/mathematical-reasoning-on-math-test-sub-category-scores.md?metric=Number%20Theory%20Accuracy>): Higher is better; unit: PERCENT.

- [Pre-algebra Accuracy](<https://www.sota2.com/research/sota/mathematical-reasoning-on-math-test-sub-category-scores.md?metric=Pre-algebra%20Accuracy>): Higher is better; unit: PERCENT.

- [Precalculus Accuracy](<https://www.sota2.com/research/sota/mathematical-reasoning-on-math-test-sub-category-scores.md?metric=Precalculus%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Algebra Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| SID-v | 94 | Not recorded | 68fa7bed1e209131cea8b588 |
| SID-c | 94 | Not recorded | 68fa7bed1e209131cea8b588 |
| DMAD | 91 | Not recorded | 68fa7bed1e209131cea8b588 |
| COT | 85 | Not recorded | 68fa7bed1e209131cea8b588 |
| IO | 85 | Not recorded | 68fa7bed1e209131cea8b588 |
| Self-Contrast | 84 | Not recorded | 68fa7bed1e209131cea8b588 |
| Self-Consistency | 75 | Not recorded | 68fa7bed1e209131cea8b588 |
| SID-v | 67 | Not recorded | 68fa7bed1e209131cea8b588 |
| SID-c | 67 | Not recorded | 68fa7bed1e209131cea8b588 |
| IO | 65 | Not recorded | 68fa7bed1e209131cea8b588 |
| SBP | 65 | Not recorded | 68fa7bed1e209131cea8b588 |
| COT | 61 | Not recorded | 68fa7bed1e209131cea8b588 |
| MAD | 61 | Not recorded | 68fa7bed1e209131cea8b588 |
| Self-Consistency | 58 | Not recorded | 68fa7bed1e209131cea8b588 |
| DMAD | 55 | Not recorded | 68fa7bed1e209131cea8b588 |
| Self-Contrast | 54 | Not recorded | 68fa7bed1e209131cea8b588 |
| SBP | 46 | Not recorded | 68fa7bed1e209131cea8b588 |

### SID-v

Algebra Accuracy: 94. Result ID: 6a16ba04fbc373592c1039fd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b588.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b588.md>)

Recorded evaluation configuration: {"Model":"GPT-OSS-20B"}. This record does not assert independent reproduction.

### SID-c

Algebra Accuracy: 94. Result ID: 6a16ba04fbc373592c1039fe.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b588.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b588.md>)

Recorded evaluation configuration: {"Model":"GPT-OSS-20B"}. This record does not assert independent reproduction.

### DMAD

Algebra Accuracy: 91. Result ID: 6a16ba04fbc373592c1039fc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b588.md>)

Recorded evaluation configuration: {"Model":"GPT-OSS-20B"}. This record does not assert independent reproduction.

### COT

Algebra Accuracy: 85. Result ID: 6a16ba04fbc373592c1039f7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b588.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e571e209131cea3f27b.md>)

Recorded evaluation configuration: {"Model":"GPT-OSS-20B"}. This record does not assert independent reproduction.

### IO

Algebra Accuracy: 85. Result ID: 6a16ba04fbc373592c1039f8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b588.md>)

Recorded evaluation configuration: {"Model":"GPT-OSS-20B"}. This record does not assert independent reproduction.

### Self-Contrast

Algebra Accuracy: 84. Result ID: 6a16ba04fbc373592c1039fb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b588.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa738f1e209131cea5d377.md>)

Recorded evaluation configuration: {"Model":"GPT-OSS-20B"}. This record does not assert independent reproduction.

### Self-Consistency

Algebra Accuracy: 75. Result ID: 6a16ba04fbc373592c1039fa.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b588.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eb61e209131cea41b33.md>)

Recorded evaluation configuration: {"Model":"GPT-OSS-20B"}. This record does not assert independent reproduction.

### SID-v

Algebra Accuracy: 67. Result ID: 6a16ba04fbc373592c1039f5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b588.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b588.md>)

Recorded evaluation configuration: {"Model":"LLaMA3.1-8B"}. This record does not assert independent reproduction.

### SID-c

Algebra Accuracy: 67. Result ID: 6a16ba04fbc373592c1039f6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b588.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b588.md>)

Recorded evaluation configuration: {"Model":"LLaMA3.1-8B"}. This record does not assert independent reproduction.

### IO

Algebra Accuracy: 65. Result ID: 6a16ba04fbc373592c1039ef.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b588.md>)

Recorded evaluation configuration: {"Model":"LLaMA3.1-8B"}. This record does not assert independent reproduction.

### SBP

Algebra Accuracy: 65. Result ID: 6a16ba04fbc373592c1039f9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b588.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72761e209131cea56856.md>)

Recorded evaluation configuration: {"Model":"GPT-OSS-20B"}. This record does not assert independent reproduction.

### COT

Algebra Accuracy: 61. Result ID: 6a16ba04fbc373592c1039ee.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b588.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e571e209131cea3f27b.md>)

Recorded evaluation configuration: {"Model":"LLaMA3.1-8B"}. This record does not assert independent reproduction.

### MAD

Algebra Accuracy: 61. Result ID: 6a16ba04fbc373592c1039f3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b588.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f6a1e209131cea456e2.md>)

Recorded evaluation configuration: {"Model":"LLaMA3.1-8B"}. This record does not assert independent reproduction.

### Self-Consistency

Algebra Accuracy: 58. Result ID: 6a16ba04fbc373592c1039f1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b588.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eb61e209131cea41b33.md>)

Recorded evaluation configuration: {"Model":"LLaMA3.1-8B"}. This record does not assert independent reproduction.

### DMAD

Algebra Accuracy: 55. Result ID: 6a16ba04fbc373592c1039f4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b588.md>)

Recorded evaluation configuration: {"Model":"LLaMA3.1-8B"}. This record does not assert independent reproduction.

### Self-Contrast

Algebra Accuracy: 54. Result ID: 6a16ba04fbc373592c1039f2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b588.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa738f1e209131cea5d377.md>)

Recorded evaluation configuration: {"Model":"LLaMA3.1-8B"}. This record does not assert independent reproduction.

### SBP

Algebra Accuracy: 46. Result ID: 6a16ba04fbc373592c1039f0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b588.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72761e209131cea56856.md>)

Recorded evaluation configuration: {"Model":"LLaMA3.1-8B"}. This record does not assert independent reproduction.

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