Penalizing Confident Predictions on Largely Perturbed Inputs Does Not Improve Out-of-Distribution Generalization in Question Answering.
Kazutoshi ShinodaSaku SugawaraAkiko AizawaPublished in: CoRR (2022)
Keyphrases
- question answering
- information retrieval
- question classification
- natural language
- cross language
- passage retrieval
- sentence retrieval
- question answering systems
- named entities
- natural language processing
- qa clef
- answering questions
- syntactic information
- artificial intelligence
- semantic roles
- natural language questions
- open domain question answering