Conversational Question Answering on Abg-CoQA Unambiguous
84.4Overlap F1IntentRL
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| IntentRLModel=4B Instruct, Output Format=Interpretation-Answer Pairs2025.11 | 84.4 | 90.9 | 40.7 | |
| SFTModel=4B Instruct, Output Format=Answer-only2025.11 | 72.2 | 77.6 | 70.6 | |
| IntentPromptModel=235B MoE Instruct, Output Format=Interpretation-Answer Pairs2025.11 | 68.9 | 80.8 | 42.1 | |
| IntentPromptModel=235B MoE Thinking, Output Format=Answer-only2025.11 | 66.1 | 85.5 | 51.5 | |
| IntentPromptModel=235B MoE Instruct, Output Format=Answer-only2025.11 | 64.7 | 76.6 | 44.7 | |
| IntentPromptModel=235B MoE Thinking, Output Format=Interpretation-Answer Pairs2025.11 | 59.6 | 82.1 | 47.2 | |
| IntentPromptModel=4B Instruct, Output Format=Interpretation-Answer Pairs2025.11 | 59.2 | 77.7 | 40.6 | |
| IntentPromptModel=4B Instruct, Output Format=Answer-only2025.11 | 56.8 | 74.8 | 40.2 | |
| IntentPromptModel=4B Thinking, Output Format=Answer-only2025.11 | 53.4 | 72.7 | 50.2 | |
| IntentPromptModel=4B Thinking, Output Format=Interpretation-Answer Pairs2025.11 | 32.5 | 60.6 | 44.6 |