Optimization Modeling on Mamo ComplexLP
67.5Accuracy (pass@1)MIRROR
Evaluation results
| Method | Links | |
|---|---|---|
| MIRRORCategory=Ours2026.02 | 67.5 | |
| SIRLCategory=Learning-based methods, Model Scale=32B2026.02 | 61.1 | |
| COECategory=Agent-based methods2026.02 | 57.63 | |
| Deepseek-v3Category=Traditional prompting2026.02 | 57.14 | |
| MIRRORCategory=Ours, Model Scale=30B2026.02 | 52.7 | |
| ORMindCategory=Agent-based methods2026.02 | 52.22 | |
| OptiMUSCategory=Agent-based methods2026.02 | 45.81 | |
| Backbone modelCategory=Traditional prompting2026.02 | 44.83 | |
| LLMOPTCategory=Learning-based methods, Model Scale=14B2026.02 | 44.08 | |
| ORLMCategory=Learning-based methods, Model Scale=8B2026.02 | 37.4 | |
| OptMATHCategory=Learning-based methods, Model Scale=7B2026.02 | 34.12 | |
| MiniOptCategory=Learning-based methods, Model Scale=14B2026.02 | 33.65 | |
| qwen3-30BCategory=Traditional prompting, Model Scale=30B2026.02 | 30.54 |