Eunsun Jou [ɯnsʌ̹n ʤu] / 주은선

Hello!

I am an Assistant Professor at The School of Digital Humanities and Computational Social Sciences at the Korea Advanced Institute of Science and Technology (KAIST).

As a linguist and syntactician, I explore the principles that underlie complex phenomena involving the structure of natural language. Recently, I have growing interests in the biological underpinnings of these principles, as well as the role of linguistic knowledge in effective and safe use of large language models.

I received my Ph.D. in linguistics at the Department of Linguistics and Philosophy at MIT in 2024. Before my Ph.D, I studied for a B.A. and M.A. in linguistics at the Department of Linguistics at Seoul National University. You can reach me at eunsunjou @ kaist.ac.kr .

CV

You can find my CV here.

The Language and Cognition Lab is hiring interns!

Theoretical syntax and semantics, with methodological range

We study the structure and meaning of natural language, couched within state-of-the-art theoretical development. Uncovering the universal principles and parameters of natural language is the fundamental basis of linguistics and linguistically aligned research. As we tackle our research questions, we aim for both theoretical precision as well as quantitative rigour. This may involve using quantitative methods such as psycholinguistic experiments and corpus analysis.

Developmentally plausible language models

Large language models are getting better at not only speaking like humans, but also at capturing structural phenomena that were deemed inaccessible to them a few years ago. However, this achievement sits on training data that is orders of magnitude larger than what the average human is exposed to during their childhood. Can we use developmentally plausible language models to probe the human language faculty without extreme data and resource usage?

Linguistically informed AI ethics

Our current project funded by the National Research Foundation of Korea aims to develop a framework for detecting implicit bias in LLM output. Unlike explicit bias, implicit bias is harder to detect and appears in subtle ways. We aim to provide a framework for detecting them based on linguistic cues such as coreference resolution. We also aspire to provide a solution that is generalizable to lower-resource languages.

Publications

Articles and Theses

2026. A successive-cyclic dependent case account of Korean case-marked adverbials. Natural Language and Linguistic Theory 44. DOI: https://doi.org/10.1007/s11049-026-09717-4
2024. Honorification as Agree in Korean and beyond. Glossa: a journal of general linguistics 9(1). DOI: https://doi.org/10.16995/glossa.9565
2024. Structural case on adjuncts. Doctoral Dissertation. MIT. [open file in new tab]
2018. Embedded topicalization in Korean factive complement clauses: An experimental approach. MA Thesis. Seoul National University. [open file in new tab]

Conference Proceedings

2024. Positional restriction on case assignment: Evidence from Korean nominal adverbials. Proceedings of WAFL 17.
2023. An economy-based amendment to learning hidden structure with Robust Interpretive Parsing. Proceedings of the 2022 Annual Meeting on Phonology. DOI: https://doi.org/10.3765/amp.v10i0.5420
2018. Embedded topicalization in Korean factive complement clauses: An experimental approach. Proceedings of Japanese/Korean Linguistics 26.
2017. The distribution and interpretation of the Korean topic marker nun - a feature-inheritance approach. Proceedings of the Linguistic Society of Korea.

Presentations

Invited talks/lectures

2026. Can large language models learn why-in-situ island violations? Invited talk at the Korean Society of Language and Information.
2026. The performative hypothesis and meta-questions in Korean. Joint winter conference of the Korean Generative Grammar Circle and The Society of Modern Grammar.
2025. Case-marked adverbials and successive-cyclic Dependent Case. Guest lecture for seminar on Korean Linguistics. University of Illinois, Urbana-Champaign.
2025. Case-marked adverbials and the timing of case evaluation. Invited talk at the Comparative Syntax, Semantics, and Language Acquisition Workshop. Nanzan University Center for Linguistics.
2025. Revisiting Successive-cyclic Dependent Case in the context of case-marked adverbials. Invited talk at the Korean Generative Grammar Circle Annual Workshop.

Conference Presentations

2025. A successive-cyclic dependent case account of adverbial case alternation in Korean passives. TEAL 14.
2025. Korean nonactive suffixes HI and eci are realizations of little v. LSA 2025.
2023. Positional restriction on case assignment: Evidence from Korean nominal adverbials. WAFL 17.
2023. Case marking of Korean nominal adverbials correlates with subject position. SICOGG 25.
2022. An economy-based amendment to Robust Interpretive Parsing with the GLA. AMP 2022.
2022. Korean addressee honoriﬁcation as cyclic agree at Force: Comparison with Magahi. GLOW 45.
2019. Embedded topicalization in factive complement clauses. 93rd meeting of the Linguistic Society of America.
2018. Embedded topicalization in Korean factive complement clauses: An experimental approach. Japanese/Korean Linguistics 26.

Teaching

Spring 2026. [KAIST HSS20060/DHS20060] Introduction to Linguistics.
Spring 2026. [KAIST HSS39940] Special core topics in humanities: Introduction to Syntax.
Spring 2025. [MIT 24.952] Advanced Syntax. Co-instructors: Elise Newman, David Pesestsky
Spring 2025. [MIT 24.S95] Special Seminar: Advanced Workshop in Syntax and Semantics. Co-instructor: Danny Fox