Optimization Modeling on BWOR
82.9AROR-LLM-Agent
Evaluation results
| Method | Links | |
|---|---|---|
| OR-LLM-AgentBackbone=DeepSeek-R12026.02 | 82.9 | |
| Grok-42026.02 | 80.5 | |
| Grok42026.02 | 80.5 | |
| R2C with reflectionReflection mechanism=Enabled, Maximum reflection iterations=32026.02 | 79.2 | |
| GPT-52026.02 | 78 | |
| GPT52026.02 | 78 | |
| R2CReflection mechanism=Disabled2026.02 | 75.6 | |
| R2CBackbone=deepseek-v3.22026.02 | 75.6 | |
| Standard PromptBackbone=deepseek-v3.22026.02 | 63.4 | |
| GPT-42026.02 | 40.2 | |
| ORLMBackbone=LLaMa3-8B2026.02 | 29.3 |