Understanding and Meeting Practitioner Needs When Measuring Representational Harms Caused by LLM-Based Systems

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Harvey, Emma, Sheng, Emily, Blodgett, Su Lin, Chouldechova, Alexandra, Garcia-Gathright, Jean, Olteanu, Alexandra, Wallach, Hanna
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915327346999296
author Harvey, Emma
Sheng, Emily
Blodgett, Su Lin
Chouldechova, Alexandra
Garcia-Gathright, Jean
Olteanu, Alexandra
Wallach, Hanna
author_facet Harvey, Emma
Sheng, Emily
Blodgett, Su Lin
Chouldechova, Alexandra
Garcia-Gathright, Jean
Olteanu, Alexandra
Wallach, Hanna
contents The NLP research community has made publicly available numerous instruments for measuring representational harms caused by large language model (LLM)-based systems. These instruments have taken the form of datasets, metrics, tools, and more. In this paper, we examine the extent to which such instruments meet the needs of practitioners tasked with evaluating LLM-based systems. Via semi-structured interviews with 12 such practitioners, we find that practitioners are often unable to use publicly available instruments for measuring representational harms. We identify two types of challenges. In some cases, instruments are not useful because they do not meaningfully measure what practitioners seek to measure or are otherwise misaligned with practitioner needs. In other cases, instruments - even useful instruments - are not used by practitioners due to practical and institutional barriers impeding their uptake. Drawing on measurement theory and pragmatic measurement, we provide recommendations for addressing these challenges to better meet practitioner needs.
format Preprint
id arxiv_https___arxiv_org_abs_2506_04482
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Understanding and Meeting Practitioner Needs When Measuring Representational Harms Caused by LLM-Based Systems
Harvey, Emma
Sheng, Emily
Blodgett, Su Lin
Chouldechova, Alexandra
Garcia-Gathright, Jean
Olteanu, Alexandra
Wallach, Hanna
Computers and Society
Computation and Language
The NLP research community has made publicly available numerous instruments for measuring representational harms caused by large language model (LLM)-based systems. These instruments have taken the form of datasets, metrics, tools, and more. In this paper, we examine the extent to which such instruments meet the needs of practitioners tasked with evaluating LLM-based systems. Via semi-structured interviews with 12 such practitioners, we find that practitioners are often unable to use publicly available instruments for measuring representational harms. We identify two types of challenges. In some cases, instruments are not useful because they do not meaningfully measure what practitioners seek to measure or are otherwise misaligned with practitioner needs. In other cases, instruments - even useful instruments - are not used by practitioners due to practical and institutional barriers impeding their uptake. Drawing on measurement theory and pragmatic measurement, we provide recommendations for addressing these challenges to better meet practitioner needs.
title Understanding and Meeting Practitioner Needs When Measuring Representational Harms Caused by LLM-Based Systems
topic Computers and Society
Computation and Language
url https://arxiv.org/abs/2506.04482