Traveling Salesperson Problem (Constructive) on TSP-Constructive N=100 (val)
7.726Mean Tour LengthOptimal
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Optimal2026.05 | 7.726 | 0 | |
| AHD Agent w/SRModel=Qwen3-4B-Instruct-2507, Eval=100, Cost ($)=0.0432026.05 | 8.795 | 13.855 | |
| CALMModel=Qwen3-4B-Instruct-2507, Eval=150, Cost ($)=0.0332026.05 | 8.852 | 14.597 | |
| AHD AgentModel=Qwen3-4B-Instruct-2507, Eval=12.8, Cost ($)=0.0102026.05 | 9.044 | 17.067 | |
| ReEvoModel=DeepSeek-V4-Flash, Eval=100, Cost ($)=0.0592026.05 | 9.093 | 17.702 | |
| EOHModel=GPT-4o, Eval=100, Cost ($)=0.2142026.05 | 9.127 | 18.136 | |
| MCTS-AHDModel=GPT-4o, Eval=100, Cost ($)=0.0732026.05 | 9.271 | 19.998 | |
| MCTS-AHDModel=DeepSeek-V4-Flash, Eval=100, Cost ($)=0.0422026.05 | 9.345 | 20.944 | |
| AHD AgentModel=DeepSeek-V4-Flash, Eval=18.4, Cost ($)=0.0542026.05 | 9.448 | 22.285 | |
| EOHModel=DeepSeek-V4-Flash, Eval=100, Cost ($)=0.0272026.05 | 9.484 | 22.759 | |
| AHD AgentModel=GPT-4o, Eval=4.6, Cost ($)=0.1232026.05 | 9.558 | 23.727 | |
| ReEvoModel=GPT-4o, Eval=100, Cost ($)=0.3922026.05 | 9.583 | 24.054 | |
| Baseline heuristic2026.05 | 9.586 | 24.094 |