2026-09-19T00:15:02Zhttps://digital.lib.washington.edu/server/oai/request

oai:digital.lib.washington.edu:1773/482822026-02-15T22:31:06Zcom_1773_4888col_1773_4936

Steinert-Threlkeld, Shane Barnes, Megan 2022-01-26T23:25:26Z 2022-01-26T23:25:26Z 2022-01-26 2021 Barnes_washington_0250O_23738.pdf http://hdl.handle.net/1773/48282 Thesis (Master's)--University of Washington, 2021 This paper investigates whether biasing natural language models toward tree-compositional structure and systematic token representation can improve performance on tasks that require the use of function words. The method used treats tree-structure as latent and thus requires no gold parse labels. Results show that across four function-word-focused NLI probing tasks, tree-compositional models perform as well as LSTMs, but lag behind BERT to varying degrees between tasks. Context-dependent behavior of tree-compositional models highlights a potential weakness of the architecture in the absence of grounding information. application/pdf en_US CC BY Linguistics Computer science Linguistics Latent Compositional Representations for English Function Word Comprehension Thesis Open Access open.access