Text-to-SQL on Spider Lite 2.0 (test)
46.6Execution Accuracy (EX)PExA
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| PExALLM=GPT-o32026.04 | 46.6 | 49.9 | 5.55 | |
| AgenticDataLLM=GPT-o32026.04 | 44.5 | — | — | |
| ReFoRCELLM=GPT-o32026.04 | 36.2 | 39.5 | 5.44 | |
| Spider-AgentLLM=GPT-o32026.04 | 26.2 | 28.7 | 5.9 |