Operations Research Modeling and Solving on MAMO Complex
68Accuracy (pass@1)LLMOPT
Evaluation results
| Method | Links | |
|---|---|---|
| LLMOPTType=Fine-tuning2026.04 | 68 | |
| OptMATHType=Fine-tuning2026.04 | 54.1 | |
| NEDTreeType=Context-based2026.04 | 49.52 | |
| ORllmAgentType=Prompt-based2026.04 | 47.86 | |
| Deepseek-R1Type=Reasoning2026.04 | 47.39 | |
| OptimusType=Prompt-based2026.04 | 44.08 | |
| GPT-o4-miniType=Reasoning2026.04 | 43.12 | |
| Chain-of-ExpertsType=Prompt-based2026.04 | 42.18 | |
| Deepseek-V3Type=Non-reasoning2026.04 | 40.28 | |
| ORLM(best)Type=Fine-tuning2026.04 | 37.4 | |
| GPT-4oType=Non-reasoning2026.04 | 33.17 |