The Zoorna Institute
for Language, AI and Society
We are a non-profit research institute working at the intersection of language and AI to address societal challenges. Our mission is to strengthen and spur research in low-resource languages, specializing on the languages of the Middle East and the Caucasus, and to apply our solutions for social good.
Syntax matters. So does Society.
About the Zoorna Institute
Founded in 2024, the Zoorna Institute is a research organization dedicated to advancing linguistically informed AI and exploring the role of language in society. Our work bridges theoretical linguistics, natural language processing, and social inquiry—with a particular focus on low-resource languages from the Middle East and the Caucasus.
We engage in original research and interdisciplinary collaborations to deepen our understanding of language structure, semantics, and variation. At the same time, we apply computational methods to analyze cultural narratives, media discourse, and the sociopolitical impact of language technologies. Whether developing corpora, designing formal grammars, or studying algorithmic bias, our goal is to build more inclusive, context-aware systems rooted in linguistic insight.
Our Work
We conduct both foundational and applied research across a range of linguistic and AI domains. Our efforts are rooted in deep theoretical inquiry and carried through in real-world applications.
Our Work
Recent Publications
& Presentations
We contribute to ongoing research and dialogue in linguistics, NLP, and AI for social impact.
AI-Enabled Narrative Analytics for Persian and Kurdish
Presented at the 4th North American Conference on Iranian Linguistics
University of Toronto Mississauga. May 2025.
Narratives are foundational to human expression across cultures. They are in the stories we tell, in folktales, news reports, memoirs, podcasts and visual media. The linguist Bill Labov describes the Narrative as "a recounting of things that have happened, involving a sequence of events meaningfully connected in a temporal and often causal relation, typically structured with a beginning, middle, and end". We used prompt engineering to develop Large Language Models or LLMs that identify the structural elements of a narrative--in other words, the system automatically extracts the information from a text to answer who did what to whom, where and when and why. We found that LLMs perform quite well on this task for Persian and Sorani Kurdish, especially in inferring implicit information and discontinuous elements, without requiring the integration of NLP pipeline components or structured resources such as WordNet or a Treebank. However, these systems are inconsistent (especially for Kurdish) and don't perform as well in complex analyses such as coreference resolution. For researchers working on endangered or minority languages, this finding opens exciting doors.
We are a young research institute. This section will expand as our research and outreach efforts continue. Please come back soon to see our progress as we build our research portfolio and our team.