Visual Riddles: a Commonsense and World Knowledge Challenge for Large Vision and Language Models.
Nitzan Bitton GuettaAviv SlobodkinAviya MaimonEliya HabbaRoyi RassinYonatan BittonIdan SzpektorAmir GlobersonYuval EloviciPublished in: CoRR (2024)
Keyphrases
- language model
- world knowledge
- language modeling
- n gram
- document retrieval
- probabilistic model
- knowledge base
- bag of words
- feature generation
- retrieval model
- information retrieval
- knowledge sources
- test collection
- noun phrases
- background knowledge
- visual information
- query terms
- vector space model
- computer vision
- query expansion
- pseudo relevance feedback
- document representation
- natural language text
- bayesian networks
- visual features
- information retrieval systems
- retrieval effectiveness
- low level
- action recognition