How Inclusively do LMs Perceive Social and Moral Norms?

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Galarnyk, Michael, Shah, Agam, Guhathakurta, Dipanwita, Nandigam, Poojitha, Chava, Sudheer
Format:	Preprint
Published:	2025
Subjects:	Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

_version_	1866913796048551936
author	Galarnyk, Michael Shah, Agam Guhathakurta, Dipanwita Nandigam, Poojitha Chava, Sudheer
author_facet	Galarnyk, Michael Shah, Agam Guhathakurta, Dipanwita Nandigam, Poojitha Chava, Sudheer
contents	This paper discusses and contains offensive content. Language models (LMs) are used in decision-making systems and as interactive assistants. However, how well do these models making judgements align with the diversity of human values, particularly regarding social and moral norms? In this work, we investigate how inclusively LMs perceive norms across demographic groups (e.g., gender, age, and income). We prompt 11 LMs on rules-of-thumb (RoTs) and compare their outputs with the existing responses of 100 human annotators. We introduce the Absolute Distance Alignment Metric (ADA-Met) to quantify alignment on ordinal questions. We find notable disparities in LM responses, with younger, higher-income groups showing closer alignment, raising concerns about the representation of marginalized perspectives. Our findings highlight the importance of further efforts to make LMs more inclusive of diverse human values. The code and prompts are available on GitHub under the CC BY-NC 4.0 license.
format	Preprint
id	arxiv_https___arxiv_org_abs_2502_02696
institution	arXiv
publishDate	2025
record_format	arxiv
spellingShingle	How Inclusively do LMs Perceive Social and Moral Norms? Galarnyk, Michael Shah, Agam Guhathakurta, Dipanwita Nandigam, Poojitha Chava, Sudheer Computation and Language This paper discusses and contains offensive content. Language models (LMs) are used in decision-making systems and as interactive assistants. However, how well do these models making judgements align with the diversity of human values, particularly regarding social and moral norms? In this work, we investigate how inclusively LMs perceive norms across demographic groups (e.g., gender, age, and income). We prompt 11 LMs on rules-of-thumb (RoTs) and compare their outputs with the existing responses of 100 human annotators. We introduce the Absolute Distance Alignment Metric (ADA-Met) to quantify alignment on ordinal questions. We find notable disparities in LM responses, with younger, higher-income groups showing closer alignment, raising concerns about the representation of marginalized perspectives. Our findings highlight the importance of further efforts to make LMs more inclusive of diverse human values. The code and prompts are available on GitHub under the CC BY-NC 4.0 license.
title	How Inclusively do LMs Perceive Social and Moral Norms?
topic	Computation and Language
url	https://arxiv.org/abs/2502.02696