ChatGPT Prompting Cannot Estimate Predictive Uncertainty in High-Resource Languages.

Martino Pelucchi Matias Valdenegro-Toro

Published in: CoRR (2023)

Keyphrases

expressive power
wide range
estimation algorithm
database
case study
language independent
accurate estimation
language identification
data sets
neural network
incomplete information
uncertain data
resource constraints
context free grammars