Reposted by @oldbayes.bsky.social
“Selecting Shots for Demographic Fairness in Few-Shot Learning with LLMs”
We evaluate different example selection methods (semantic- and demographic-based) finding that semantic similarity works best for fairness, but it’s not the best all the time!
aclanthology.org/2024.nlp4pi-...
aclanthology.org
Selecting Shots for Demographic Fairness in Few-Shot Learning with Large Language Models
Carlos Alejandro Aguirre, Kuleen Sasse, Isabel Alyssa Cachola, Mark Dredze. Proceedings of the Third Workshop on NLP for Positive Impact. 2024.