Optimization Modeling on MAMO-E
88.34AccuracyORLM (SFT)
Evaluation results
| Method | Links | |
|---|---|---|
| ORLM (SFT)Paradigm=Training-based, Base Model=Qwen2.5-7B, Solver=Gurobi2026.04 | 88.34 | |
| EVOM (GRPO)Paradigm=Training-based, Base Model=Qwen2.5-7B, Solver=Gurobi2026.04 | 88.19 | |
| OpenAI o1Paradigm=Prompting-based, Solver=Gurobi2026.04 | 87.6 | |
| GPT-4o (CoE)Paradigm=Prompting-based, Solver=Gurobi2026.04 | 85.7 | |
| OptiMUSParadigm=Prompting-based, Solver=Gurobi2026.04 | 85.1 | |
| GPT-4o (CoT)Paradigm=Prompting-based, Solver=Gurobi2026.04 | 84.8 | |
| GPT-4o (Standard)Paradigm=Prompting-based, Solver=Gurobi2026.04 | 84.3 | |
| DeepSeek-R1-0528Paradigm=Prompting-based, Solver=Gurobi2026.04 | 79.45 | |
| Base ModelParadigm=Training-based, Base Model=Qwen2.5-7B, Solver=Gurobi2026.04 | 65.95 |