Sense Extension
Word sense extension (WSE) focuses on how words acquire new meanings beyond their established senses, a crucial aspect of language understanding and generation. Current research employs various techniques, including adapting existing language models and developing novel analogy-based methods, to predict and model these meaning extensions, often leveraging existing lexical resources like WordNet while addressing limitations of fine-grained sense inventories. Improved WSE models promise advancements in natural language processing tasks such as word sense disambiguation and figurative language understanding, ultimately leading to more robust and nuanced computational models of human language.
Papers
September 10, 2024
March 24, 2024