Selective Conformal Uncertainty (SConU) filters test questions with a conformal outlier test before prediction, and SConU-Pro refines that test, aiming to preserve coverage in large language model question answering.
The probability of each true class (label) is denoted as ˆf (Xi)Y∗ i
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
SConU: Selective Conformal Uncertainty in Large Language Models
Selective Conformal Uncertainty (SConU) filters test questions with a conformal outlier test before prediction, and SConU-Pro refines that test, aiming to preserve coverage in large language model question answering.