Game of 24 on 4nu 137 games (test)
90.51AccuracyXoT
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| XoTLLM Backbone=GPT-3.5, r=32023.11 | 90.51 | 1.72 | 95.94 | |
| XoTLLM Backbone=GPT-3.5, r=22023.11 | 88.32 | 1.58 | 93.87 | |
| XoTLLM Backbone=GPT-4, r=32023.11 | 85.4 | 1.78 | 92.48 | |
| XoTLLM Backbone=GPT-4, r=22023.11 | 83.94 | 1.57 | 89.63 | |
| XoTLLM Backbone=GPT-3.5, r=12023.11 | 79.56 | 1.39 | 92.15 | |
| XoTLLM Backbone=GPT-4, r=12023.11 | 74.45 | 1.38 | 88.2 | |
| MCTS2023.11 | 62.77 | — | — | |
| ToTLLM Backbone=GPT-4, b=32023.11 | 60.58 | 39.83 | — | |
| ToTLLM Backbone=GPT-4, b=12023.11 | 34.31 | 23.5 | — | |
| GoTLLM Backbone=GPT-4, k=12023.11 | 10.95 | 7 | — | |
| ToTLLM Backbone=GPT-3.5, b=32023.11 | 10.22 | 43.96 | — | |
| IOLLM Backbone=GPT-42023.11 | 10.22 | 1 | — | |
| IOLLM Backbone=GPT-3.52023.11 | 6.57 | 1 | — | |
| ToTLLM Backbone=GPT-3.5, b=12023.11 | 5.84 | 22.11 | — | |
| CoTLLM Backbone=GPT-42023.11 | 4.38 | 1 | — | |
| CoT-SCLLM Backbone=GPT-42023.11 | 4.38 | 10 | — | |
| GoTLLM Backbone=GPT-3.5, k=12023.11 | 2.92 | 7 | — | |
| CoTLLM Backbone=GPT-3.52023.11 | 2.19 | 1 | — | |
| CoT-SCLLM Backbone=GPT-3.52023.11 | 2.19 | 10 | — | |
| LLaMA-2-13B2023.11 | 2.19 | — | — |