Multi-turn Text-to-SQL on SParC Out-of-domain
77.4Execution Accuracy (EX)
[Long-Horizon] Warm-Start SFT + RL (Outcome + Process)
Updated 1mo ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| 77.4 | 69.1 | ||
| 76 | 69 | ||
2025.10 | 75.1 | 68.9 | |
| 74.5 | 68 | ||
| 74 | 64 | ||
| 74 | 64.4 | ||
| 73.7 | 62.4 | ||
2025.10 | 73.4 | 64 | |
| 73 | 66.2 | ||
2025.10 | 71.7 | 65.1 | |
| 71.3 | 62.7 | ||
2025.10 | 70.6 | 62.3 | |
| 70 | 61.5 | ||
| 67.3 | 58.1 | ||
2025.10 | 61.9 | 56.1 | |
2025.10 | 57.9 | 48.5 |