# Mathematical Reasoning on Olympiad (Pass@1, Pass@32)

Source: [Mathematical Reasoning on Olympiad (Pass@1, Pass@32)](<https://www.sota2.com/research/sota/mathematical-reasoning-on-olympiad-pass-at-1-pass-at-32>)

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

Data updated: 2026-02-26T05:57:30.778Z

Evaluation of model performance on mathematical reasoning problems using the Olympiad dataset, specifically measured by Pass@1 and Pass@32 metrics.

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) | Olympiad |
| Metric | Pass@1 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 18 of 18 |
| 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

- [Pass@1](<https://www.sota2.com/research/sota/mathematical-reasoning-on-olympiad-pass-at-1-pass-at-32.md?metric=Pass%401>): Higher is better; unit: PERCENT.

- [Pass@32](<https://www.sota2.com/research/sota/mathematical-reasoning-on-olympiad-pass-at-1-pass-at-32.md?metric=Pass%4032>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Pass@1 | Result date | Report paper |
| --- | --- | --- | --- |
| DynaMO | 31.6 | Not recorded | 699d307c6017c4460ffab3b7 |
| Fork Tokens | 30.9 | Not recorded | 699d307c6017c4460ffab3b7 |
| W-REINFORCE | 30.9 | Not recorded | 699d307c6017c4460ffab3b7 |
| Entropy Advantages | 30.4 | Not recorded | 699d307c6017c4460ffab3b7 |
| Entropy Loss | 30.2 | Not recorded | 699d307c6017c4460ffab3b7 |
| KL-COV | 30.2 | Not recorded | 699d307c6017c4460ffab3b7 |
| Clip-COV | 29.5 | Not recorded | 699d307c6017c4460ffab3b7 |
| GRPO | 28.6 | Not recorded | 699d307c6017c4460ffab3b7 |
| Clip-Higher | 28.1 | Not recorded | 699d307c6017c4460ffab3b7 |
| DynaMO | 27.2 | Not recorded | 699d307c6017c4460ffab3b7 |
| Clip-Higher | 26.4 | Not recorded | 699d307c6017c4460ffab3b7 |
| Entropy Advantages | 25.9 | Not recorded | 699d307c6017c4460ffab3b7 |
| Clip-COV | 25.8 | Not recorded | 699d307c6017c4460ffab3b7 |
| GRPO | 25.6 | Not recorded | 699d307c6017c4460ffab3b7 |
| Fork Tokens | 25.5 | Not recorded | 699d307c6017c4460ffab3b7 |
| Entropy Loss | 25.4 | Not recorded | 699d307c6017c4460ffab3b7 |
| KL-COV | 25.4 | Not recorded | 699d307c6017c4460ffab3b7 |
| W-REINFORCE | 24.4 | Not recorded | 699d307c6017c4460ffab3b7 |

### DynaMO

Pass@1: 31.6. Result ID: 699e3b735698665ec4f83825.

[Reporting paper](<https://www.sota2.com/research/paper/699d307c6017c4460ffab3b7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/699d307c6017c4460ffab3b7.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-Math-7B"}. This record does not assert independent reproduction.

### Fork Tokens

Pass@1: 30.9. Result ID: 699e3b735698665ec4f83820.

[Reporting paper](<https://www.sota2.com/research/paper/699d307c6017c4460ffab3b7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79031e209131cea7ba9b.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-Math-7B"}. This record does not assert independent reproduction.

### W-REINFORCE

Pass@1: 30.9. Result ID: 699e3b735698665ec4f83824.

[Reporting paper](<https://www.sota2.com/research/paper/699d307c6017c4460ffab3b7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7ba71e209131cea89a9e.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-Math-7B"}. This record does not assert independent reproduction.

### Entropy Advantages

Pass@1: 30.4. Result ID: 699e3b735698665ec4f83821.

[Reporting paper](<https://www.sota2.com/research/paper/699d307c6017c4460ffab3b7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a921e209131cea83f9d.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-Math-7B"}. This record does not assert independent reproduction.

### Entropy Loss

Pass@1: 30.2. Result ID: 699e3b735698665ec4f8381f.

[Reporting paper](<https://www.sota2.com/research/paper/699d307c6017c4460ffab3b7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a921e209131cea83f9d.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-Math-7B"}. This record does not assert independent reproduction.

### KL-COV

Pass@1: 30.2. Result ID: 699e3b735698665ec4f83823.

[Reporting paper](<https://www.sota2.com/research/paper/699d307c6017c4460ffab3b7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b98b.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-Math-7B"}. This record does not assert independent reproduction.

### Clip-COV

Pass@1: 29.5. Result ID: 699e3b735698665ec4f83822.

[Reporting paper](<https://www.sota2.com/research/paper/699d307c6017c4460ffab3b7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b98b.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-Math-7B"}. This record does not assert independent reproduction.

### GRPO

Pass@1: 28.6. Result ID: 699e3b735698665ec4f8381d.

[Reporting paper](<https://www.sota2.com/research/paper/699d307c6017c4460ffab3b7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73101e209131cea5a395.md>)

[Implementation](<https://github.com/deepseek-ai/DeepSeek-Math>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-Math-7B"}. This record does not assert independent reproduction.

### Clip-Higher

Pass@1: 28.1. Result ID: 699e3b735698665ec4f8381e.

[Reporting paper](<https://www.sota2.com/research/paper/699d307c6017c4460ffab3b7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78a81e209131cea79ac6.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-Math-7B"}. This record does not assert independent reproduction.

### DynaMO

Pass@1: 27.2. Result ID: 699e3b735698665ec4f8381c.

[Reporting paper](<https://www.sota2.com/research/paper/699d307c6017c4460ffab3b7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/699d307c6017c4460ffab3b7.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-Math-1.5B"}. This record does not assert independent reproduction.

### Clip-Higher

Pass@1: 26.4. Result ID: 699e3b735698665ec4f83815.

[Reporting paper](<https://www.sota2.com/research/paper/699d307c6017c4460ffab3b7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78a81e209131cea79ac6.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-Math-1.5B"}. This record does not assert independent reproduction.

### Entropy Advantages

Pass@1: 25.9. Result ID: 699e3b735698665ec4f83818.

[Reporting paper](<https://www.sota2.com/research/paper/699d307c6017c4460ffab3b7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a921e209131cea83f9d.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-Math-1.5B"}. This record does not assert independent reproduction.

### Clip-COV

Pass@1: 25.8. Result ID: 699e3b735698665ec4f83819.

[Reporting paper](<https://www.sota2.com/research/paper/699d307c6017c4460ffab3b7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b98b.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-Math-1.5B"}. This record does not assert independent reproduction.

### GRPO

Pass@1: 25.6. Result ID: 699e3b735698665ec4f83814.

[Reporting paper](<https://www.sota2.com/research/paper/699d307c6017c4460ffab3b7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73101e209131cea5a395.md>)

[Implementation](<https://github.com/deepseek-ai/DeepSeek-Math>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-Math-1.5B"}. This record does not assert independent reproduction.

### Fork Tokens

Pass@1: 25.5. Result ID: 699e3b735698665ec4f83817.

[Reporting paper](<https://www.sota2.com/research/paper/699d307c6017c4460ffab3b7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79031e209131cea7ba9b.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-Math-1.5B"}. This record does not assert independent reproduction.

### Entropy Loss

Pass@1: 25.4. Result ID: 699e3b735698665ec4f83816.

[Reporting paper](<https://www.sota2.com/research/paper/699d307c6017c4460ffab3b7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a921e209131cea83f9d.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-Math-1.5B"}. This record does not assert independent reproduction.

### KL-COV

Pass@1: 25.4. Result ID: 699e3b735698665ec4f8381a.

[Reporting paper](<https://www.sota2.com/research/paper/699d307c6017c4460ffab3b7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79001e209131cea7b98b.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-Math-1.5B"}. This record does not assert independent reproduction.

### W-REINFORCE

Pass@1: 24.4. Result ID: 699e3b735698665ec4f8381b.

[Reporting paper](<https://www.sota2.com/research/paper/699d307c6017c4460ffab3b7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7ba71e209131cea89a9e.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-Math-1.5B"}. This record does not assert independent reproduction.

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