Do Language Models Know Their Slang? Queer Slang Understanding in User-Generated Content

arXiv:2608.04847 · cs.CL · Submitted 2026-08-05 · Read on arXiv

Arianna Denitto, Beatrice Savoldi

cs.CL

Submitted: 2026-08-05

Code: https://github.com/ariadne28/Slang-Q

Project page: https://kittensyzygy.github.io/polari-dictionary/home.html

License: http://creativecommons.org/licenses/by/4.0/

The gist: Despite its cultural relevance and diffusion, queer slang remains underrepresented in Natural Language Processing research.

Terminology

Abstract

Despite its cultural relevance and diffusion, queer slang remains underrepresented in Natural Language Processing research. Towards addressing this gap, we introduce Slang-Q, a manually curated dataset of naturally user-generated English sentences paired with queer slang terms and reference definitions, built upon a newly constructed taxonomy of 118 queer terms. We use this resource to conduct a first exploratory evaluation of language models on their ability to understand and define queer slang under varying prompting conditions. Slang-Q is intended as a basis for studying how current models handle sensitive, community-specific language and whether they can provide accurate and reliable information about such forms of identity and linguistic expression.

Sources

Related papers