Combinatorial Optimization on CVRP (test)
87.2AccuracyGuided+RAG
Evaluation Results
| Method | Links | |
|---|---|---|
| Guided+RAGAgent Pair=GPT+Qwen2026.02 | 87.2 | |
| Guided+RAGAgent Pair=GPT+Llama2026.02 | 85.9 | |
| Guided+RAGAgent Pair=Avg.2026.02 | 81.3 | |
| Guided+RAGAgent Pair=Claude+Llama2026.02 | 77.7 | |
| GuidedAgent Pair=GPT+Llama2026.02 | 76.9 | |
| Guided+RAGAgent Pair=Claude+Qwen2026.02 | 74.5 | |
| Chain-of-ThoughtAgent Pair=Claude+Qwen2026.02 | 73.4 | |
| GuidedAgent Pair=GPT+Qwen2026.02 | 73.1 | |
| GuidedAgent Pair=Claude+Qwen2026.02 | 72.6 | |
| GuidedAgent Pair=Avg.2026.02 | 71.6 | |
| Chain-of-ThoughtAgent Pair=GPT+Qwen2026.02 | 69 | |
| No-GuidanceAgent Pair=GPT+Qwen2026.02 | 66.8 | |
| Chain-of-ThoughtAgent Pair=Claude+Llama2026.02 | 66.6 | |
| Chain-of-ThoughtAgent Pair=Avg.2026.02 | 66.6 | |
| No-GuidanceAgent Pair=Claude+Llama2026.02 | 65.5 | |
| No-GuidanceAgent Pair=Avg.2026.02 | 65 | |
| No-GuidanceAgent Pair=Claude+Qwen2026.02 | 64.8 | |
| GuidedAgent Pair=Claude+Llama2026.02 | 63.8 | |
| No-GuidanceAgent Pair=GPT+Llama2026.02 | 62.8 | |
| Chain-of-ThoughtAgent Pair=GPT+Llama2026.02 | 57.3 |