List your company

Question Answering on HotpotQA (F1 Score Only)

59.93F1 ScoreSynWorld
49.821252.445655.0757.6944Apr 4, 2025

Evaluation results

MethodLinks
SynWorldModel=GPT-4-turbo2025.0459.93
SynWorldModel=Qwen-long2025.0459.91
SynWorldModel=Qwen2-72B-Instruct2025.0458.7
EasyToolModel=Qwen-long2025.0458.34
EasyToolModel=GPT-4-turbo2025.0458.19
DRAFTModel=GPT-4-turbo2025.0457.71
DRAFTModel=Qwen2-72B-Instruct2025.0457.57
Self-RefineModel=Qwen-long2025.0456.1
Self-RefineModel=GPT-4-turbo2025.0455.85
EasyToolModel=Qwen2-72B-Instruct2025.0454.94
ReActModel=GPT-4-turbo2025.0454.61
DRAFTModel=Qwen-long2025.0453.23
Self-RefineModel=Qwen2-72B-Instruct2025.0452.59
ReActModel=Qwen-long2025.0452
ReActModel=Qwen2-72B-Instruct2025.0450.21