The invited talk by Prof. Juri Alexander Opitz on “Natural Language Processing RELIES on Linguistics” by Opitz. et al. 2025
Juri Opitz is a researcher in computational linguistics at the University of Zurich, where he works on semantic structures and search as part of an interdisciplinary research project. He completed his PhD at Heidelberg University in 2024, focusing on metrics for meaning representations and their application to NLP tasks like evaluation of parsing and generation, as well as explainable semantic search. You can find more at www.juriopitz.com.
Large Language Models (LLMs) have become capable of generating highly fluent text in certain languages, without modules specially designed to capture grammar or semantic coherence. What does this mean for the future of linguistic expertise in NLP? We highlight several aspects in which NLP (still) relies on linguistics, or where linguistic thinking can illuminate new directions. We argue our case around the acronym RELIES, which encapsulates six major facets where linguistics contributes to NLP: Resources, Evaluation, Low-resource settings, Interpretability, Explanation, and the Study of language. This list is not exhaustive, nor is linguistics the main point of reference for every effort under these themes; but at a macro level, these facets highlight the enduring importance of studying machine systems vis-a-vis systems of human language.
Presentation available “https://docs.google.com/presentation/d/1Vs05R8zEXpzaA2DLGOeL2qM1CKRe1dRNdyhiSRvA5DE/edit?slide=id.p#slide=id.p”