Attribute Alignment: Controlling Text Generation from Pre-trained Language Models.
Dian YuZhou YuKenji SagaePublished in: EMNLP (Findings) (2021)
Keyphrases
- language model
- pre trained
- text generation
- natural language generation
- language modeling
- information retrieval
- training data
- natural language
- document retrieval
- n gram
- retrieval model
- probabilistic model
- speech recognition
- language modelling
- theorem prover
- training examples
- query expansion
- test collection
- statistical language models
- control signals
- document ranking
- information retrieval systems
- smoothing methods
- neural network
- translation model
- multi modal
- small number
- machine learning