Does Refusal Training in LLMs Generalize to the Past Tense?

Maksym Andriushchenko Nicolas Flammarion

Published in: CoRR (2024)

Keyphrases

real world
training process
training samples
training set
neural network
training phase
training algorithm
test set
feedforward neural networks
database
supervised learning
small number
object recognition
search algorithm
multiscale
feature extraction
information retrieval
real time