MemSum: Extractive Summarization of Long Documents Using Multi-Step Episodic Markov Decision Processes.
Nianlong GuElliott AshRichard H. R. HahnloserPublished in: ACL (1) (2022)
Keyphrases
- markov decision processes
- extractive summarization
- multi step
- state space
- optimal policy
- finite state
- transition matrices
- dynamic programming
- policy iteration
- reinforcement learning
- decision theoretic planning
- average cost
- text summarization
- average reward
- infinite horizon
- knn
- k nearest neighbor
- markov decision process
- information retrieval systems
- least squares
- action sets
- reward function
- reinforcement learning algorithms
- action space
- information retrieval
- information extraction
- model based reinforcement learning
- natural language
- real time dynamic programming