Improving Human Text Comprehension through Semi-Markov CRF-based Neural Section Title Generation.
Sebastian GehrmannSteven LayneFranck DernoncourtPublished in: NAACL-HLT (1) (2019)
Keyphrases
- semi markov
- conditional random fields
- keywords
- text generation
- graphical models
- crf model
- hidden markov models
- network architecture
- pairwise
- maximum entropy
- text mining
- neural network
- generation process
- generative model
- probabilistic model
- human generated
- information extraction
- text retrieval
- contextual features
- free text
- named entity recognition
- database
- human experts
- semantic segmentation
- information retrieval
- random fields
- superpixels
- natural language generation
- bayesian networks
- machine learning
- reading comprehension
- bio inspired
- segmentation method
- human subjects