Ebhaam at SemEval-2023 Task 1: A CLIP-Based Approach for Comparing Cross-modality and Unimodality in Visual Word Sense Disambiguation.
Zeinab TaghaviParsa Haghighi NaeiniMohammad Ali Sadraei JavaheriSoroush GooranEhsaneddin AsgariHamid Reza RabieeHossein SametiPublished in: SemEval@ACL (2023)
Keyphrases
- word sense disambiguation
- wordnet
- natural language processing
- machine translation
- information extraction
- feature generation
- semantic relatedness
- visual features
- unsupervised word sense disambiguation
- wide coverage
- lexical knowledge
- linguistic knowledge
- sense disambiguation
- word sense
- lexical information
- low level
- machine learning methods
- co occurrence
- highly ambiguous
- machine learning
- semantic relations
- part of speech
- visual perception
- feature set
- domain knowledge