Text-to-SQL Parsing on Ambrosia Ambiguous subset (test)
82.4RecallIntentRL
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| IntentRLModel=4B Instruct, Output Format=Interpretation-Answer Pairs2025.11 | 82.4 | 77.5 | 74.1 | |
| SFTModel=4B Instruct, Output Format=Answer-only2025.11 | 72.7 | 75.1 | 66.4 | |
| SFTModel=4B Instruct, Output Format=Interpretation-Answer Pairs2025.11 | 68.5 | 75.6 | 58.8 | |
| IntentPromptModel=235B MoE Instruct, Output Format=Interpretation-Answer Pairs2025.11 | 46.2 | 48.9 | 18.3 | |
| IntentPromptModel=235B MoE Thinking, Output Format=Interpretation-Answer Pairs2025.11 | 37.3 | 37.8 | 12.2 | |
| IntentPromptModel=235B MoE Instruct, Output Format=Answer-only2025.11 | 35.2 | 53.3 | 8.4 | |
| IntentPromptModel=235B MoE Thinking, Output Format=Answer-only2025.11 | 28.2 | 32.8 | 5.3 | |
| IntentPromptModel=4B Thinking, Output Format=Answer-only2025.11 | 27.4 | 49.1 | 2.3 | |
| IntentPromptModel=4B Instruct, Output Format=Answer-only2025.11 | 23.8 | 48.1 | 2.3 | |
| IntentPromptModel=4B Instruct, Output Format=Interpretation-Answer Pairs2025.11 | 22.5 | 30.8 | 3.8 | |
| IntentPromptModel=4B Thinking, Output Format=Interpretation-Answer Pairs2025.11 | 16.9 | 23.2 | 3.1 |