Reward-Augmented Decoding: Efficient Controlled Text Generation With a Unidirectional Reward Model.

Haikang Deng Colin Raffel

Published in: EMNLP (2023)

Keyphrases

experimental data
computational model
probabilistic model
probability distribution
formal model
reinforcement learning
high level
management system
theoretical analysis
statistical model
data mining
objective function
prior knowledge
data sets
machine learning
theoretical framework
conceptual model
neural network