The Effects of Demographic Instructions on LLM Personas

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: de Paula, Angel Felipe Magnossão, Culpepper, J. Shane, Moffat, Alistair, Cherumanal, Sachin Pathiyan, Scholer, Falk, Trippas, Johanne
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910950561415168
author de Paula, Angel Felipe Magnossão
Culpepper, J. Shane
Moffat, Alistair
Cherumanal, Sachin Pathiyan
Scholer, Falk
Trippas, Johanne
author_facet de Paula, Angel Felipe Magnossão
Culpepper, J. Shane
Moffat, Alistair
Cherumanal, Sachin Pathiyan
Scholer, Falk
Trippas, Johanne
contents Social media platforms must filter sexist content in compliance with governmental regulations. Current machine learning approaches can reliably detect sexism based on standardized definitions, but often neglect the subjective nature of sexist language and fail to consider individual users' perspectives. To address this gap, we adopt a perspectivist approach, retaining diverse annotations rather than enforcing gold-standard labels or their aggregations, allowing models to account for personal or group-specific views of sexism. Using demographic data from Twitter, we employ large language models (LLMs) to personalize the identification of sexism.
format Preprint
id arxiv_https___arxiv_org_abs_2505_11795
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle The Effects of Demographic Instructions on LLM Personas
de Paula, Angel Felipe Magnossão
Culpepper, J. Shane
Moffat, Alistair
Cherumanal, Sachin Pathiyan
Scholer, Falk
Trippas, Johanne
Information Retrieval
Social media platforms must filter sexist content in compliance with governmental regulations. Current machine learning approaches can reliably detect sexism based on standardized definitions, but often neglect the subjective nature of sexist language and fail to consider individual users' perspectives. To address this gap, we adopt a perspectivist approach, retaining diverse annotations rather than enforcing gold-standard labels or their aggregations, allowing models to account for personal or group-specific views of sexism. Using demographic data from Twitter, we employ large language models (LLMs) to personalize the identification of sexism.
title The Effects of Demographic Instructions on LLM Personas
topic Information Retrieval
url https://arxiv.org/abs/2505.11795