Improving Factual Consistency of Text Summarization by Adversarially Decoupling Comprehension and Embellishment Abilities of LLMs.
Huawen FengYan FanXiong LiuTing-En LinZekun YaoYuchuan WuFei HuangYongbin LiQianli MaPublished in: CoRR (2023)
Keyphrases
- text summarization
- named entity recognition
- natural language processing
- information extraction
- multi document summarization
- question answering
- text segments
- automatic text summarization
- extractive summarization
- query expansion
- automatic summarization
- keyword extraction
- arabic language
- named entities
- model selection
- temporal expressions
- lexical chains
- conditional random fields
- document structure
- semi supervised
- learning process
- artificial intelligence