Current Projects
Explore our list of active projects
Narrative Analytics
Persian and Kurdish
Exploring how large language models (LLMs) can perform narrative analysis in Persian and Sorani Kurdish, this project investigates the models' capabilities in event extraction, timeline construction, and understanding implicit information, even in low-resource language settings.
AI and Education
AI-powered language learning tools
Developing AI-driven tools to facilitate language learning, this initiative focuses on creating adaptive learning platforms that cater to the unique challenges of low-resource languages, enhancing accessibility and engagement for learners worldwide. Current pilot is for Persian and Armenian.
Tajik Transliteration
Bridging Persian dialects across scripts and borders
This project focuses on building a bidirectional transliteration system between Tajik Persian (Cyrillic script) and Iranian Persian (Perso-Arabic script). We are comparing machine learning models with generative AI approaches to evaluate accuracy, linguistic consistency, and adaptability. The project supports cross-script communication and resource development for underrepresented Persian varieties across Central and Western Asia.
The Art of Taarof
Modeling ritual politeness in Persian
Taarof, the Persian system of ritual politeness, is deeply embedded in sociocultural norms. In collaboration with Brock University, we are examining how Taarof can be computationally modeled and understood, with applications for culturally aware AI systems, language learning tools, and pragmatic annotation frameworks.
Verbal Reduplication
Eastern Armenian
This linguistic research project explores how verbal reduplication in Eastern Armenian encodes nuanced distinctions in event plurality, distribution, and intensity. Using examples like կտրել (ktrel, "to cut") vs. կտրտել (ktrtel, "to chop (repeatedly)"), we analyze the morphosyntactic patterns and semantic interpretations of reduplicated verb forms in Armenian.
Scientometrics
Exploring NLP and Linguistics Publications
This project uses bibliometrics and social network analysis to compare research trends in NLP and theoretical linguistics over the past decade. By analyzing publication data, citation patterns, authorship networks, and thematic shifts, we aim to understand the evolving relationship—and divergence—between computational and linguistic research communities.