Optimization Modeling on Mamo EasyLP
94.6Accuracy (pass@1)SIRL
Evaluation results
| Method | Links | |
|---|---|---|
| SIRLCategory=Learning-based methods, Model Scale=32B2026.02 | 94.6 | |
| MiniOptCategory=Learning-based methods, Model Scale=14B2026.02 | 90.8 | |
| LLMOPTCategory=Learning-based methods, Model Scale=14B2026.02 | 89.53 | |
| COECategory=Agent-based methods2026.02 | 87.93 | |
| MIRRORCategory=Ours2026.02 | 87.3 | |
| MIRRORCategory=Ours, Model Scale=30B2026.02 | 86.9 | |
| qwen3-30BCategory=Traditional prompting, Model Scale=30B2026.02 | 85.28 | |
| OptiMUSCategory=Agent-based methods2026.02 | 85.28 | |
| Backbone modelCategory=Traditional prompting2026.02 | 85.07 | |
| Deepseek-v3Category=Traditional prompting2026.02 | 84.66 | |
| OptMATHCategory=Learning-based methods, Model Scale=7B2026.02 | 84.2 | |
| ORLMCategory=Learning-based methods, Model Scale=8B2026.02 | 82.3 | |
| ORMindCategory=Agent-based methods2026.02 | 81.19 |