Puzzle Solving on 8-Puzzle
95.8AccuracyXoT
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| XoTLLM=GPT-4, Revision count=32023.11 | 95.8 | 1.61 | 62.22 | |
| XoTLLM=GPT-4, Revision count=22023.11 | 94.96 | 1.55 | 58.91 | |
| XoTLLM=GPT-4, Revision count=12023.11 | 93.28 | 1.48 | 55.66 | |
| XoTLLM=GPT-3.5, Revision count=32023.11 | 63.03 | 2.29 | 42.6 | |
| XoTLLM=GPT-3.5, Revision count=12023.11 | 59.66 | 1.5 | 41.09 | |
| XoTLLM=GPT-3.5, Revision count=22023.11 | 59.66 | 1.92 | 42.18 | |
| MCTSLLM=GPT-3.52023.11 | 51.26 | — | — | |
| MCTSLLM=GPT-42023.11 | 51.26 | — | — | |
| ToTLLM=GPT-4, Branching factor=32023.11 | 13.45 | 54.13 | — | |
| CoT-SCLLM=GPT-42023.11 | 8.4 | 10 | — | |
| CoTLLM=GPT-42023.11 | 7.56 | 1 | — | |
| ToTLLM=GPT-3.5, Branching factor=32023.11 | 6.72 | 55.86 | — | |
| ToTLLM=GPT-3.5, Branching factor=12023.11 | 5.88 | 31.76 | — | |
| GoTLLM=GPT-3.5, k=12023.11 | 3.36 | 19 | — | |
| ToTLLM=GPT-4, Branching factor=12023.11 | 3.36 | 27.49 | — | |
| GoTLLM=GPT-4, k=12023.11 | 3.36 | 19 | — | |
| IOLLM=GPT-42023.11 | 1.68 | 1 | — | |
| CoT-SCLLM=GPT-3.52023.11 | 0.84 | 10 | — | |
| IOLLM=GPT-3.52023.11 | 0 | 1 | — | |
| CoTLLM=GPT-3.52023.11 | 0 | 1 | — | |
| LLaMA-2-13BLLM=GPT-3.52023.11 | 0 | — | — | |
| LLaMA-2-13BLLM=GPT-42023.11 | 0 | — | — |