Gradient Ascent Post-training Enhances Language Model Generalization.

Dongkeun Yoon Joel Jang Sungdong Kim Minjoon Seo

Published in: CoRR (2023)

Keyphrases

language model
gradient ascent
language modeling
n gram
information retrieval
cross entropy
probabilistic model
test collection
retrieval model
document retrieval
query expansion
expectation maximization
mixture model
training set
translation model
prior knowledge
machine learning