List your company

Workflow Extraction on ABCD

58.55Macro ScoreQA-CoT
34.858841.009447.1653.3106Feb 24, 2025

Evaluation results

MethodLinks
QA-CoTBackbone=GPT-4o2025.0258.5556.2910.6
QA-CoT + ReflectBackbone=Opus2025.0256.8456.6511.54
QA-CoTBackbone=Sonnet2025.0256.2853.4411.49
QA-CoT + ReflectBackbone=GPT-4o2025.0255.4555.3910.73
QA-CoTBackbone=Opus2025.0252.3655.7511.92
ReflectBackbone=Sonnet2025.0251.849.4610.57
QA-CoT + ReflectBackbone=Sonnet2025.0251.1154.6710.56
ReflectBackbone=Opus2025.0250.8945.8610.47
QA-CoTBackbone=Gemini2025.0248.0946.48.4
BasicBackbone=GPT-4o2025.0246.7444.968.69
EnsembleBackbone=Sonnet2025.0246.4742.449.29
EnsembleBackbone=GPT-4o2025.0246.1345.328.66
BasicBackbone=Sonnet2025.0245.68418.86
BasicBackbone=Opus2025.0245.1540.829.99
ReflectBackbone=GPT-4o2025.0244.4342.098.51
QA-CoT + ReflectBackbone=Gemini2025.0243.146.228.36
BasicBackbone=Gemini2025.0242.836.517.89
EnsembleBackbone=Opus2025.0241.834.358.51
PlanBackbone=Gemini2025.024138.126.75
EnsembleBackbone=Gemini2025.0238.7138.678.42
PlanBackbone=Sonnet2025.0237.5634.717.39
PlanBackbone=GPT-4o2025.0237.2938.37.71
ReflectBackbone=Gemini2025.0237.1435.618.51
PlanBackbone=Opus2025.0235.7732.917.9