Saved in:
Bibliographic Details
Main Authors: Sheffield, William, Misra, Kanishka, Pyatkin, Valentina, Deo, Ashwini, Mahowald, Kyle, Li, Junyi Jessy
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2506.04534
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918046306664448
author Sheffield, William
Misra, Kanishka
Pyatkin, Valentina
Deo, Ashwini
Mahowald, Kyle
Li, Junyi Jessy
author_facet Sheffield, William
Misra, Kanishka
Pyatkin, Valentina
Deo, Ashwini
Mahowald, Kyle
Li, Junyi Jessy
contents Discourse particles are crucial elements that subtly shape the meaning of text. These words, often polyfunctional, give rise to nuanced and often quite disparate semantic/discourse effects, as exemplified by the diverse uses of the particle "just" (e.g., exclusive, temporal, emphatic). This work investigates the capacity of LLMs to distinguish the fine-grained senses of English "just", a well-studied example in formal semantics, using data meticulously created and labeled by expert linguists. Our findings reveal that while LLMs exhibit some ability to differentiate between broader categories, they struggle to fully capture more subtle nuances, highlighting a gap in their understanding of discourse particles.
format Preprint
id arxiv_https___arxiv_org_abs_2506_04534
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Is It JUST Semantics? A Case Study of Discourse Particle Understanding in LLMs
Sheffield, William
Misra, Kanishka
Pyatkin, Valentina
Deo, Ashwini
Mahowald, Kyle
Li, Junyi Jessy
Computation and Language
Artificial Intelligence
Discourse particles are crucial elements that subtly shape the meaning of text. These words, often polyfunctional, give rise to nuanced and often quite disparate semantic/discourse effects, as exemplified by the diverse uses of the particle "just" (e.g., exclusive, temporal, emphatic). This work investigates the capacity of LLMs to distinguish the fine-grained senses of English "just", a well-studied example in formal semantics, using data meticulously created and labeled by expert linguists. Our findings reveal that while LLMs exhibit some ability to differentiate between broader categories, they struggle to fully capture more subtle nuances, highlighting a gap in their understanding of discourse particles.
title Is It JUST Semantics? A Case Study of Discourse Particle Understanding in LLMs
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2506.04534