Task Planning on SheetCopilot 10% subset 1.0
85Exec@1GPT-3.5-Turbo
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| GPT-3.5-TurboToken limit=4096, Data proportion=10%2023.05 | 85 | 45 | 2 | 4.5 | |
| ClaudeToken limit=4096, Data proportion=10%2023.05 | 80 | 40 | 1.5 | 4.4 | |
| GPT-4Token limit=4096, Data proportion=10%2023.05 | 65 | 55 | 1.33 | 2 |