SciNLI: A Corpus for Natural Language Inference on Scientific Text.
Mobashir SadatCornelia CarageaPublished in: CoRR (2022)
Keyphrases
- natural language
- natural language text
- scientific papers
- broad coverage
- natural language generation
- open domain
- text data
- supervised machine learning
- information extraction
- text generation
- text retrieval
- natural language processing
- plain text
- semantic representation
- text corpus
- world knowledge
- named entity disambiguation
- text corpora
- scientific literature
- newspaper articles
- information retrieval
- natural language sentences
- human language
- manually annotated
- question answering
- semantic interpretation
- english words
- keywords
- lexical semantics
- topic segmentation
- linguistic patterns
- linguistic analysis
- linguistic information
- lexical features
- meaning representations
- semantic analysis
- free text
- text mining
- knowledge representation
- document corpus
- semantic markup
- scientific documents
- machine learning
- information extraction systems
- training corpus
- text collections
- text documents
- entity extraction
- statistical machine translation
- sentence level
- recognizing textual entailment
- language processing
- bayesian networks