# Multilingual Knowledge Evaluation on Global-MMLU Low

Source: [Multilingual Knowledge Evaluation on Global-MMLU Low](<https://www.sota2.com/research/sota/multilingual-knowledge-evaluation-on-global-mmlu-low>)

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

Data updated: 2026-06-08T06:32:02.864Z

Evaluation of model performance on the Global-MMLU Low dataset variant, focusing on multilingual knowledge capabilities, measured by accuracy.

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 | Multilingual knowledge evaluation |
| Dataset (including any stated split) | Global-MMLU Low |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 12 of 12 |
| 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/multilingual-knowledge-evaluation-on-global-mmlu-low.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| GRPO | 52.31 | Not recorded | 6a264c8590966ade919be956 |
| Baseline | 52.19 | Not recorded | 6a264c8590966ade919be956 |
| CPT + SFT | 51.19 | Not recorded | 6a264c8590966ade919be956 |
| CPT + GRPO | 50 | Not recorded | 6a264c8590966ade919be956 |
| CPT | 49.19 | Not recorded | 6a264c8590966ade919be956 |
| SFT | 46.75 | Not recorded | 6a264c8590966ade919be956 |
| GRPO | 32 | Not recorded | 6a264c8590966ade919be956 |
| Baseline | 31.79 | Not recorded | 6a264c8590966ade919be956 |
| CPT + SFT | 30.88 | Not recorded | 6a264c8590966ade919be956 |
| SFT | 30.32 | Not recorded | 6a264c8590966ade919be956 |
| CPT + GRPO | 29.61 | Not recorded | 6a264c8590966ade919be956 |
| CPT | 29.32 | Not recorded | 6a264c8590966ade919be956 |

### GRPO

Accuracy: 52.31. Result ID: 6a266179fbc373592c226c72.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

Recorded evaluation configuration: {"Backbone":"Qwen-2.5-7B"}. This record does not assert independent reproduction.

### Baseline

Accuracy: 52.19. Result ID: 6a266179fbc373592c226c6e.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

Recorded evaluation configuration: {"Backbone":"Qwen-2.5-7B"}. This record does not assert independent reproduction.

### CPT + SFT

Accuracy: 51.19. Result ID: 6a266179fbc373592c226c71.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

Recorded evaluation configuration: {"Backbone":"Qwen-2.5-7B"}. This record does not assert independent reproduction.

### CPT + GRPO

Accuracy: 50. Result ID: 6a266179fbc373592c226c73.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

Recorded evaluation configuration: {"Backbone":"Qwen-2.5-7B"}. This record does not assert independent reproduction.

### CPT

Accuracy: 49.19. Result ID: 6a266179fbc373592c226c6f.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

Recorded evaluation configuration: {"Backbone":"Qwen-2.5-7B"}. This record does not assert independent reproduction.

### SFT

Accuracy: 46.75. Result ID: 6a266179fbc373592c226c70.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

Recorded evaluation configuration: {"Backbone":"Qwen-2.5-7B"}. This record does not assert independent reproduction.

### GRPO

Accuracy: 32. Result ID: 6a266179fbc373592c226c6c.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

Recorded evaluation configuration: {"Backbone":"OLMo-2-1124-7B"}. This record does not assert independent reproduction.

### Baseline

Accuracy: 31.79. Result ID: 6a266179fbc373592c226c68.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

Recorded evaluation configuration: {"Backbone":"OLMo-2-1124-7B"}. This record does not assert independent reproduction.

### CPT + SFT

Accuracy: 30.88. Result ID: 6a266179fbc373592c226c6b.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

Recorded evaluation configuration: {"Backbone":"OLMo-2-1124-7B"}. This record does not assert independent reproduction.

### SFT

Accuracy: 30.32. Result ID: 6a266179fbc373592c226c6a.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

Recorded evaluation configuration: {"Backbone":"OLMo-2-1124-7B"}. This record does not assert independent reproduction.

### CPT + GRPO

Accuracy: 29.61. Result ID: 6a266179fbc373592c226c6d.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

Recorded evaluation configuration: {"Backbone":"OLMo-2-1124-7B"}. This record does not assert independent reproduction.

### CPT

Accuracy: 29.32. Result ID: 6a266179fbc373592c226c69.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c8590966ade919be956.md>)

Recorded evaluation configuration: {"Backbone":"OLMo-2-1124-7B"}. This record does not assert independent reproduction.

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