# Multilingual Reward Modeling on MM-Eval Avg. 18 lang

Source: [Multilingual Reward Modeling on MM-Eval Avg. 18 lang](<https://www.sota2.com/research/sota/multilingual-reward-modeling-on-mm-eval-avg-18-lang>)

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

Data updated: 2026-03-12T07:53:43.970Z

Evaluation of model performance on the Multilingual Reward Modeling task using the MM-Eval average subset covering 18 languages, 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 Reward Modeling |
| Dataset (including any stated split) | MM-Eval Avg. 18 lang |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 14 of 14 |
| 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-reward-modeling-on-mm-eval-avg-18-lang.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| DIBJudge-Qwen3-8B | 87.53 | Not recorded | 69b24878d030084c22e2f85c |
| mR3-Qwen3-8B | 85.29 | Not recorded | 69b24878d030084c22e2f85c |
| DIBJudge-Qwen3-4B | 85.16 | Not recorded | 69b24878d030084c22e2f85c |
| mR3-Qwen3-4B | 82.55 | Not recorded | 69b24878d030084c22e2f85c |
| Qwen3-8B | 82.2 | Not recorded | 69b24878d030084c22e2f85c |
| Qwen3-4B | 80.85 | Not recorded | 69b24878d030084c22e2f85c |
| Gemini-2.5-Flash | 77.47 | Not recorded | 69b24878d030084c22e2f85c |
| Nemotron-Multi-49B | 76.31 | Not recorded | 69b24878d030084c22e2f85c |
| Think-as-Locals 7B | 72.95 | Not recorded | 69b24878d030084c22e2f85c |
| GPT-4o | 71.85 | Not recorded | 69b24878d030084c22e2f85c |
| M-PROMETHEUS 7B | 69.38 | Not recorded | 69b24878d030084c22e2f85c |
| Qwen2.5-7B-Instruct | 65.64 | Not recorded | 69b24878d030084c22e2f85c |
| M-PROMETHEUS 3B | 64.17 | Not recorded | 69b24878d030084c22e2f85c |
| Qwen2.5-3B-Instruct | 57.99 | Not recorded | 69b24878d030084c22e2f85c |

### DIBJudge-Qwen3-8B

Accuracy: 87.53. Result ID: 69b270606dbb7075b95ccab2.

[Reporting paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-8B"}. This record does not assert independent reproduction.

### mR3-Qwen3-8B

Accuracy: 85.29. Result ID: 69b270606dbb7075b95ccaaf.

[Reporting paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bc51e209131cea8a58d.md>)

Recorded evaluation configuration: {"Model Category":"Multilingual Open Reward Models"}. This record does not assert independent reproduction.

### DIBJudge-Qwen3-4B

Accuracy: 85.16. Result ID: 69b270606dbb7075b95ccab1.

[Reporting paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-4B"}. This record does not assert independent reproduction.

### mR3-Qwen3-4B

Accuracy: 82.55. Result ID: 69b270606dbb7075b95ccaae.

[Reporting paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bc51e209131cea8a58d.md>)

Recorded evaluation configuration: {"Model Category":"Multilingual Open Reward Models"}. This record does not assert independent reproduction.

### Qwen3-8B

Accuracy: 82.2. Result ID: 69b270606dbb7075b95ccaaa.

[Reporting paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Model Category":"General Open Models"}. This record does not assert independent reproduction.

### Qwen3-4B

Accuracy: 80.85. Result ID: 69b270606dbb7075b95ccaa9.

[Reporting paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Model Category":"General Open Models"}. This record does not assert independent reproduction.

### Gemini-2.5-Flash

Accuracy: 77.47. Result ID: 69b270606dbb7075b95ccaa6.

[Reporting paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c311e209131cea8cef5.md>)

Recorded evaluation configuration: {"Model Category":"Proprietary Models"}. This record does not assert independent reproduction.

### Nemotron-Multi-49B

Accuracy: 76.31. Result ID: 69b270606dbb7075b95ccaab.

[Reporting paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa706f1e209131cea4af71.md>)

[Implementation](<https://github.com/i-Eval/FairEval>)

Recorded evaluation configuration: {"Model Category":"Multilingual Open Reward Models"}. This record does not assert independent reproduction.

### Think-as-Locals 7B

Accuracy: 72.95. Result ID: 69b270606dbb7075b95ccab0.

[Reporting paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7ba81e209131cea89b47.md>)

Recorded evaluation configuration: {"Model Category":"Multilingual Open Reward Models"}. This record does not assert independent reproduction.

### GPT-4o

Accuracy: 71.85. Result ID: 69b270606dbb7075b95ccaa5.

[Reporting paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75681e209131cea67a98.md>)

Recorded evaluation configuration: {"Model Category":"Proprietary Models"}. This record does not assert independent reproduction.

### M-PROMETHEUS 7B

Accuracy: 69.38. Result ID: 69b270606dbb7075b95ccaad.

[Reporting paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77e71e209131cea75e93.md>)

Recorded evaluation configuration: {"Model Category":"Multilingual Open Reward Models"}. This record does not assert independent reproduction.

### Qwen2.5-7B-Instruct

Accuracy: 65.64. Result ID: 69b270606dbb7075b95ccaa8.

[Reporting paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa767e1e209131cea6d6a7.md>)

Recorded evaluation configuration: {"Model Category":"General Open Models"}. This record does not assert independent reproduction.

### M-PROMETHEUS 3B

Accuracy: 64.17. Result ID: 69b270606dbb7075b95ccaac.

[Reporting paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77e71e209131cea75e93.md>)

Recorded evaluation configuration: {"Model Category":"Multilingual Open Reward Models"}. This record does not assert independent reproduction.

### Qwen2.5-3B-Instruct

Accuracy: 57.99. Result ID: 69b270606dbb7075b95ccaa7.

[Reporting paper](<https://www.sota2.com/research/paper/69b24878d030084c22e2f85c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa767e1e209131cea6d6a7.md>)

Recorded evaluation configuration: {"Model Category":"General Open Models"}. This record does not assert independent reproduction.

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