Do language models practice what they preach? Examining language ideologies about gendered language reform encoded in LLMs

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Watson, Julia, Lee, Sophia, Beekhuizen, Barend, Stevenson, Suzanne
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929509576474624
author Watson, Julia
Lee, Sophia
Beekhuizen, Barend
Stevenson, Suzanne
author_facet Watson, Julia
Lee, Sophia
Beekhuizen, Barend
Stevenson, Suzanne
contents We study language ideologies in text produced by LLMs through a case study on English gendered language reform (related to role nouns like congressperson/-woman/-man, and singular they). First, we find political bias: when asked to use language that is "correct" or "natural", LLMs use language most similarly to when asked to align with conservative (vs. progressive) values. This shows how LLMs' metalinguistic preferences can implicitly communicate the language ideologies of a particular political group, even in seemingly non-political contexts. Second, we find LLMs exhibit internal inconsistency: LLMs use gender-neutral variants more often when more explicit metalinguistic context is provided. This shows how the language ideologies expressed in text produced by LLMs can vary, which may be unexpected to users. We discuss the broader implications of these findings for value alignment.
format Preprint
id arxiv_https___arxiv_org_abs_2409_13852
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Do language models practice what they preach? Examining language ideologies about gendered language reform encoded in LLMs
Watson, Julia
Lee, Sophia
Beekhuizen, Barend
Stevenson, Suzanne
Computation and Language
Artificial Intelligence
We study language ideologies in text produced by LLMs through a case study on English gendered language reform (related to role nouns like congressperson/-woman/-man, and singular they). First, we find political bias: when asked to use language that is "correct" or "natural", LLMs use language most similarly to when asked to align with conservative (vs. progressive) values. This shows how LLMs' metalinguistic preferences can implicitly communicate the language ideologies of a particular political group, even in seemingly non-political contexts. Second, we find LLMs exhibit internal inconsistency: LLMs use gender-neutral variants more often when more explicit metalinguistic context is provided. This shows how the language ideologies expressed in text produced by LLMs can vary, which may be unexpected to users. We discuss the broader implications of these findings for value alignment.
title Do language models practice what they preach? Examining language ideologies about gendered language reform encoded in LLMs
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2409.13852