Traveling Salesperson Problem (ACO) on TSP-ACO N=200 (val)
10.682Mean CostOptimal
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Optimal2026.05 | 10.682 | 0 | |
| AHD Agent w/SRModel=Qwen3-4B-Instruct-2507, Eval=100, Cost ($)=0.0312026.05 | 12.015 | 12.483 | |
| AHD AgentModel=Qwen3-4B-Instruct-2507, Eval=11.8, Cost ($)=0.0032026.05 | 12.039 | 12.705 | |
| AHD AgentModel=DeepSeek-V4-Flash, Eval=14.2, Cost ($)=0.0612026.05 | 12.115 | 13.415 | |
| EOHModel=GPT-4o, Eval=100, Cost ($)=0.2362026.05 | 12.218 | 14.385 | |
| ReEvoModel=DeepSeek-V4-Flash, Eval=100, Cost ($)=0.0512026.05 | 12.232 | 14.514 | |
| MCTS-AHDModel=GPT-4o, Eval=100, Cost ($)=1.0262026.05 | 12.302 | 15.274 | |
| EOHModel=DeepSeek-V4-Flash, Eval=100, Cost ($)=0.0132026.05 | 12.37 | 15.798 | |
| MCTS-AHDModel=DeepSeek-V4-Flash, Eval=100, Cost ($)=0.1042026.05 | 12.395 | 16.039 | |
| CALMModel=Qwen3-4B-Instruct-2507, Eval=150, Cost ($)=0.0692026.05 | 12.672 | 18.63 | |
| ReEvoModel=GPT-4o, Eval=100, Cost ($)=0.4732026.05 | 12.811 | 19.943 | |
| AHD AgentModel=GPT-4o, Eval=7, Cost ($)=0.0582026.05 | 13.299 | 24.307 | |
| Baseline heuristic2026.05 | 15.304 | 43.276 |