Share your thoughts, 1 month free Claude Pro on usSee more

Conversational Financial Question Answering on ConvFinQA (test)

78.81Accuracy

Program-of-Thoughts (PoT)

Updated 1mo ago

Evaluation Results

Method	Links
Program-of-Thoughts (PoT) 2024.02		78.81
EEDP 2024.02		77.91
Chain-of-Thought (CoT) 2024.02		71.19
EEDP 2024.02		63.42
Direct 2024.02		63.1
EEDP 2024.02		61.75
Program-of-Thoughts (PoT) 2024.02		61.19
Chain-of-Thought (CoT) 2024.02		48.45
Chain-of-Thought (CoT) 2024.02		48.33
Chain-of-Thought (CoT) 2024.02		46.08
Program-of-Thoughts (PoT) 2024.02		40.14
EEDP 2024.02		39.42
Chain-of-Thought (CoT) 2024.02		38
Direct 2024.02		37.62
EEDP 2024.02		36.57
EEDP 2024.02		32.77
Program-of-Thoughts (PoT) 2024.02		29.92
Chain-of-Thought (CoT) 2024.02		29.45
Decomposers 2024.02		22.32
Direct 2024.02		21.61
Direct 2024.02		20.19
Decomposers 2024.02		18.76
Program-of-Thoughts (PoT) 2024.02		14.48
Direct 2024.02		12.35
Decomposers 2024.02		11.16
Decomposers 2024.02		10.5
Decomposers 2024.02		10.35
Program-of-Thoughts (PoT) 2024.02		8.78
Decomposers 2024.02		7.83
Direct 2024.02		3.8