Question Answering on HotpotQA (F1 Score Only)
59.93F1 ScoreSynWorld
Evaluation Results
| Method | Links | |
|---|---|---|
| SynWorldModel=GPT-4-turbo2025.04 | 59.93 | |
| SynWorldModel=Qwen-long2025.04 | 59.91 | |
| SynWorldModel=Qwen2-72B-Instruct2025.04 | 58.7 | |
| EasyToolModel=Qwen-long2025.04 | 58.34 | |
| EasyToolModel=GPT-4-turbo2025.04 | 58.19 | |
| DRAFTModel=GPT-4-turbo2025.04 | 57.71 | |
| DRAFTModel=Qwen2-72B-Instruct2025.04 | 57.57 | |
| Self-RefineModel=Qwen-long2025.04 | 56.1 | |
| Self-RefineModel=GPT-4-turbo2025.04 | 55.85 | |
| EasyToolModel=Qwen2-72B-Instruct2025.04 | 54.94 | |
| ReActModel=GPT-4-turbo2025.04 | 54.61 | |
| DRAFTModel=Qwen-long2025.04 | 53.23 | |
| Self-RefineModel=Qwen2-72B-Instruct2025.04 | 52.59 | |
| ReActModel=Qwen-long2025.04 | 52 | |
| ReActModel=Qwen2-72B-Instruct2025.04 | 50.21 |