Second Thoughts are Best: Learning to Re-Align With Human Values from Text Edits.
Ruibo LiuChenyan JiaGe ZhangZiyu ZhuangTony X. LiuSoroush VosoughiPublished in: CoRR (2023)
Keyphrases
- learning process
- human learning
- learning systems
- neural network
- reinforcement learning
- database
- prior knowledge
- online learning
- learning tasks
- case study
- attribute values
- data sets
- language acquisition
- background knowledge
- active learning
- natural language
- keywords
- learning algorithm
- information retrieval
- machine learning