PoBRL: Optimizing Multi-Document Summarization by Blending Reinforcement Learning Policies.
Andy SuDifei SuJohn M. MulveyH. Vincent PoorPublished in: CoRR (2021)
Keyphrases
- multi document summarization
- reinforcement learning
- optimal policy
- reward function
- state space
- dynamic programming
- mutual reinforcement
- document summarization
- markov decision processes
- multidocument summarization
- text summarization
- manifold ranking
- knowledge base
- cross document
- supervised learning
- sentence extraction
- sentence similarity
- shallow semantic
- single document summarization