Ohio State is in the process of revising websites and program materials to accurately reflect compliance with the law. While this work occurs, language referencing protected class status or other activities prohibited by Ohio Senate Bill 1 may still appear in some places. However, all programs and activities are being administered in compliance with federal and state law.

Sara Court

Sara Court

Sara-Malta

Contact Information

Doctoral Student
she / 她 / hija / elle / ella

Google Map

Advisor: Micha Elsner

My research interests lie at the intersection of computational modeling, morphological typology, and low-resource machine learning. 

I'm committed to collaborative, ethical, and responsible AI development, particularly for underrepresented and minoritized language varieties. I'm thinking a lot these days about how we can leverage LLMs and other generative AI models to shift old paradigms and make the development of language technologies more equitable and accessible. 

I specialize in morphological typology and low-resource languages in contact, which is to say, I strive to build models that can interface with a wide range of linguistic structures as they're actually used in real life -- with all the messiness, complexity, and rich variability that make them so beautifully human! 

 

Current Projects:

  • AI/ML for Collaborative Language Technology Development
    • Responsibly adapting LLMs in low-resource domains
    • Science outreach and AI Literacy: "LLMs in the Real World" (preprint)
  • Computational Modeling of the Maltese Lexicon
    • Information theoretic analysis of Maltese morphology in contact 
    • Multi-agent RAG pipeline for low-resource Machine Translation (MT)

Past Projects:

  • CogSciDIY language science outreach for high school students
  • Active learning methods for lexical and morphological annotation
  • Countability and classifiers in Taiwanese Hokkien
  • Quantitative analysis of Mandarin morphology

Presentations:

Publications:

Links to all publications are available via my Google Scholar page.