Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Rationale Generation on GSM8K (train)
Loading...
98.9
Rationale Validity Rate
TRICE without CV
79.452
84.501
89.55
94.599
Nov 28, 2023
Rationale Validity Rate
Updated 1mo ago
Evaluation Results
Method
Method
Links
Rationale Validity Rate
TRICE without CV
Initialization=Bootstr...
2023.11
98.9
TRICE with CV
Initialization=Bootstr...
2023.11
98.8
TRICE with CV (not bootstrapped)
Initialization=Random...
2023.11
98.4
STaR
Initialization=Bootstr...
2023.11
80.2
Feedback
Search any
task
Search any
task