Towards Robust NLG Bias Evaluation with Syntactically-diverse Prompts.

Arshiya Aggarwal Jiao Sun Nanyun Peng

Published in: EMNLP (Findings) (2022)

Keyphrases

natural language generation
wide variety
database systems
natural language
evaluation model
computationally efficient
evaluation methods
highly accurate
learning processes
comparative evaluation
parameter tuning
evaluation criteria
evaluation metrics
data sets
multi agent
feature selection
real world