Please join us on Friday, February 13, 2026, from 12:30 PM to 1:30 PM for the in-person ISP Forum in SENSQ 5317. We will be featuring Janet Liu . 

Lunch and refreshments will be provided, starting at noon.

Beyond Fluency: Linguistically-Grounded Perspectives on Making Sense of LLMs 

Abstract: While large language models (LLMs) are now able to produce fluent text with ease, often giving the impression of deep linguistic competence, this fluency can mask a shallow grasp of the linguistic structures, particularly at the discourse level, that organize meaning beyond individual sentences. Human language production, by contrast, exhibits remarkable richness and variation, reflecting diverse communication styles and goals, with meaning unfolding across sentences or interactions. This offers both a challenge and an opportunity for understanding what it means for models to “make sense”. 

In this talk, I present a research program that brings together linguistic theory and NLP techniques to develop more principled, variation-aware approaches to LLM analysis and evaluation. Through probing discourse phenomena, decomposing model reasoning, and variation-aware evaluation, I aim to show both the capabilities and limitations of current LLMs. Together, I argue for integrating linguistic insights with practical challenges, contributing to the development of systems and evaluation that are more interpretable, robust, and closely aligned with human understanding. 

Bio: Yang Janet Liu is an Assistant Professor in the Department of Linguistics at the University of Pittsburgh. Her research interests involve 1) computational approaches to discourse-level linguistic phenomena across text types and their NLP applications such as summarization, 2) cross-framework discourse understanding and unifying discourse resources, and 3) addressing linguistic and domain variation in NLP. 

She co-organizes several computational linguistics and NLP workshops and shared tasks, such as the 20th Linguistic Annotation Workshop at ACL 2026, the First Workshop on Bridging NLP and Public Opinion Research at COLM 2025, and the DISRPT shared task at the Workshop on Computational Approaches to Discourse (2021, 2023, 2025). 

Before joining Pitt, she was a Postdoctoral Researcher at the MaiNLP research lab at the Center for Information and Language Processing at LMU Munich led by Prof. Dr. Barbara Plank and was affiliated with the Munich Center for Machine Learning. She received her Ph.D. in Computational Linguistics from Georgetown University. Previously, she did research internships at Spotify (2021, 2023) and Alex AI at Amazon (2020).