Learning to Maximize Mutual Information for Chain-of-Thought Distillation.
Xin ChenHanxian HuangYanjun GaoYi WangJishen ZhaoKe DingPublished in: ACL (Findings) (2024)
Keyphrases
- mutual information
- learning process
- learning algorithm
- inductive learning
- learning problems
- learning systems
- incremental learning
- knowledge acquisition
- positive examples
- learning analytics
- information theoretic
- online learning
- supervised learning
- active learning
- prior knowledge
- expert systems
- text classification
- collaborative learning
- unsupervised learning
- reinforcement learning
- case study
- image processing
- inductive inference
- database