Table Reasoning on WikiTQ (D)
0.651AccuracyPOTABLE
Evaluation Results
| Method | Links | |
|---|---|---|
| POTABLELLM=Llama-3.1-70B-Instruct2024.12 | 0.651 | |
| POTABLELLM=GPT-4o-mini2024.12 | 0.6358 | |
| Chain-of-TableLLM=Llama-3.1-70B-Instruct2024.12 | 0.6239 | |
| BinderLLM=GPT-4o-mini2024.12 | 0.592 | |
| TabSQLifyLLM=GPT-4o-mini2024.12 | 0.5687 | |
| DaterLLM=GPT-4o-mini2024.12 | 0.5676 | |
| Chain-of-TableLLM=GPT-4o-mini2024.12 | 0.5664 | |
| TabSQLifyLLM=Llama-3.1-70B-Instruct2024.12 | 0.5551 | |
| BinderLLM=Llama-3.1-70B-Instruct2024.12 | 0.5065 | |
| DaterLLM=Llama-3.1-70B-Instruct2024.12 | 0.4278 |