Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Conversational Numerical Reasoning Question Answering on ConvFinQA v1 (test)
Loading...
89.44
Execution Accuracy
Human Expert Performance
45.1984
56.6842
68.17
79.6558
Dec 14, 2022
Execution Accuracy
Program Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Execution Accuracy
Program Accuracy
Human Expert Performance
2022.12
89.44
86.34
APOLLO
Protocol=Fine-tuning,...
2022.12
78.76
77.19
APOLLO
Protocol=Fine-tuning
2022.12
76
74.56
FinQANet
Protocol=Fine-tuning,...
2022.12
68.9
68.24
T-5
Protocol=Fine-tuning,...
2022.12
58.66
57.05
GPT-2
Protocol=Fine-tuning,...
2022.12
58.19
57
General Crowd Performance
2022.12
46.9
45.52
Feedback
Search any
task
Search any
task