What's "up" with vision-language models? Investigating their struggle with spatial reasoning.
Amita KamathJack HesselKai-Wei ChangPublished in: EMNLP (2023)
Keyphrases
- spatial reasoning
- language model
- language modeling
- spatial relations
- probabilistic model
- document retrieval
- n gram
- language modelling
- speech recognition
- temporal reasoning
- query expansion
- computer vision
- mixture model
- test collection
- spatial knowledge
- retrieval model
- statistical language models
- topological relations
- directional relations
- information retrieval
- smoothing methods
- vector space model
- context sensitive
- translation model
- document ranking
- qualitative spatial reasoning
- language models for information retrieval
- pseudo relevance feedback
- relevance model
- multiscale
- spatial information
- feature vectors
- keywords
- machine learning