e-ViL: A Dataset and Benchmark for Natural Language Explanations in Vision-Language Tasks.
Maxime KayserOana-Maria CamburuLeonard SalewskiCornelius EmdeVirginie DoZeynep AkataThomas LukasiewiczPublished in: CoRR (2021)
Keyphrases
- natural language
- language processing
- natural language interface
- knowledge representation
- visually guided
- machine learning
- real time
- formal language
- linguistic knowledge
- information extraction
- natural language processing
- semantic representation
- generating explanations
- image processing
- benchmark datasets
- semantic analysis
- database
- real world
- controlled natural language
- computer vision
- natural language understanding
- conceptual graphs
- machine translation
- natural language generation
- natural language text
- multiple tasks
- knowledge base
- question answering
- service robots
- linguistic analysis
- formal languages
- human language
- causal models