Training a model to write programs that answer cohorts of similar questions consistently improves the paper's in-house cohort accuracy on most benchmarks.
Becel: Benchmark for consistency evaluation of language models
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CL 1years
2025 1verdicts
REJECT 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
CC-LEARN: Cohort-based Consistency Learning
Training a model to write programs that answer cohorts of similar questions consistently improves the paper's in-house cohort accuracy on most benchmarks.