Perturbing Inputs for Fragile Interpretations in Deep Natural Language Processing.
Sanchit SinhaHanjie ChenArshdeep SekhonYangfeng JiYanjun QiPublished in: CoRR (2021)
Keyphrases
- natural language processing
- information extraction
- natural language
- text processing
- machine learning
- named entity recognition
- computational linguistics
- wordnet
- language processing
- text mining
- text summarization
- word sense disambiguation
- semantic relations
- free text
- linguistic knowledge
- question answering
- knowledge representation
- semantic analysis
- machine translation
- part of speech
- input variables
- data sets
- named entities
- website
- artificial intelligence
- information retrieval
- feature engineering