# Mathematical Reasoning on MMK12 Source: [Mathematical Reasoning on MMK12]() Collection index: [llms.txt]() Data updated: 2026-06-09T09:43:58.514Z Evaluation of model performance on the MMK12 dataset for Mathematical Reasoning tasks, 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 | Mathematical Reasoning | | Dataset (including any stated split) | MMK12 | | Metric | Accuracy | | 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 - [Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | PRPO | 82.8 | Not recorded | 6a279e6e90966ade919bfac6 | | PAPO-D | 80.6 | Not recorded | 6a279e6e90966ade919bfac6 | | VPPO | 80.1 | Not recorded | 6a279e6e90966ade919bfac6 | | DAPO | 75.9 | Not recorded | 6a279e6e90966ade919bfac6 | | GRPO | 72.3 | Not recorded | 6a279e6e90966ade919bfac6 | | PAPO-G | 71.8 | Not recorded | 6a279e6e90966ade919bfac6 | | R1-ShareVL-7B | 70.9 | Not recorded | 6a279e6e90966ade919bfac6 | | VL-Rethinker-7B | 68.3 | Not recorded | 6a279e6e90966ade919bfac6 | | MM-Eureka-7B | 67.7 | Not recorded | 6a279e6e90966ade919bfac6 | | PRPO | 66.9 | Not recorded | 6a279e6e90966ade919bfac6 | | DAPO | 66.8 | Not recorded | 6a279e6e90966ade919bfac6 | | VPPO | 65.3 | Not recorded | 6a279e6e90966ade919bfac6 | | PAPO-D | 64.1 | Not recorded | 6a279e6e90966ade919bfac6 | | GRPO | 57.2 | Not recorded | 6a279e6e90966ade919bfac6 | | PAPO-G | 56.3 | Not recorded | 6a279e6e90966ade919bfac6 | | NoisyRollout-7B | 51 | Not recorded | 6a279e6e90966ade919bfac6 | | Qwen2.5-VL-7B | 43.3 | Not recorded | 6a279e6e90966ade919bfac6 | | Qwen2.5-VL-3B | 37.1 | Not recorded | 6a279e6e90966ade919bfac6 | ### PRPO Accuracy: 82.8. Result ID: 6a27deeafbc373592c250776. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"7B","Base Model":"Qwen2.5-VL-7B","Rollout":"8"}. This record does not assert independent reproduction. ### PAPO-D Accuracy: 80.6. Result ID: 6a27deeafbc373592c250774. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"7B","Base Model":"Qwen2.5-VL-7B","Rollout":"8"}. This record does not assert independent reproduction. ### VPPO Accuracy: 80.1. Result ID: 6a27deeafbc373592c250775. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Scale":"7B","Base Model":"Qwen2.5-VL-7B","Rollout":"8"}. This record does not assert independent reproduction. ### DAPO Accuracy: 75.9. Result ID: 6a27deeafbc373592c250772. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"7B","Base Model":"Qwen2.5-VL-7B","Rollout":"8"}. This record does not assert independent reproduction. ### GRPO Accuracy: 72.3. Result ID: 6a27deeafbc373592c250771. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Scale":"7B","Base Model":"Qwen2.5-VL-7B","Rollout":"8"}. This record does not assert independent reproduction. ### PAPO-G Accuracy: 71.8. Result ID: 6a27deeafbc373592c250773. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"7B","Base Model":"Qwen2.5-VL-7B","Rollout":"8"}. This record does not assert independent reproduction. ### R1-ShareVL-7B Accuracy: 70.9. Result ID: 6a27deeafbc373592c25076f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"7B","Rollout":"8"}. This record does not assert independent reproduction. ### VL-Rethinker-7B Accuracy: 68.3. Result ID: 6a27deeafbc373592c25076d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"7B","Rollout":"8"}. This record does not assert independent reproduction. ### MM-Eureka-7B Accuracy: 67.7. Result ID: 6a27deeafbc373592c25076c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"7B","Rollout":"8"}. This record does not assert independent reproduction. ### PRPO Accuracy: 66.9. Result ID: 6a27deeafbc373592c25076b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"3B","Base Model":"Qwen2.5-VL-3B","Rollout":"8"}. This record does not assert independent reproduction. ### DAPO Accuracy: 66.8. Result ID: 6a27deeafbc373592c250767. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"3B","Base Model":"Qwen2.5-VL-3B","Rollout":"8"}. This record does not assert independent reproduction. ### VPPO Accuracy: 65.3. Result ID: 6a27deeafbc373592c25076a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Scale":"3B","Base Model":"Qwen2.5-VL-3B","Rollout":"8"}. This record does not assert independent reproduction. ### PAPO-D Accuracy: 64.1. Result ID: 6a27deeafbc373592c250769. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"3B","Base Model":"Qwen2.5-VL-3B","Rollout":"8"}. This record does not assert independent reproduction. ### GRPO Accuracy: 57.2. Result ID: 6a27deeafbc373592c250766. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Scale":"3B","Base Model":"Qwen2.5-VL-3B","Rollout":"8"}. This record does not assert independent reproduction. ### PAPO-G Accuracy: 56.3. Result ID: 6a27deeafbc373592c250768. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"3B","Base Model":"Qwen2.5-VL-3B","Rollout":"8"}. This record does not assert independent reproduction. ### NoisyRollout-7B Accuracy: 51. Result ID: 6a27deeafbc373592c25076e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"7B","Rollout":"8"}. This record does not assert independent reproduction. ### Qwen2.5-VL-7B Accuracy: 43.3. Result ID: 6a27deeafbc373592c250770. [Reporting paper]() Recorded evaluation configuration: {"Model Scale":"7B","Rollout":"8"}. This record does not assert independent reproduction. ### Qwen2.5-VL-3B Accuracy: 37.1. Result ID: 6a27deeafbc373592c250765. [Reporting paper]() Recorded evaluation configuration: {"Model Scale":"3B","Rollout":"8"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.