Desiderata for Actionable Bias Research - Laboratoire Interdisciplinaire des Sciences du Numérique
Conference Papers Year : 2024

Desiderata for Actionable Bias Research

Abstract

The identification of stereotypical biases in NLP tools is receiving increasing attention, as corpora, metrics and mitigation techniques are being developed. These resources are instrumental to make progress towards harm mitigation. Building on these early successes of bias research, we present some desiderata to move the field forward in three actionable directions: increasing the visibility of bias evaluations, widening studies beyond gender bias and engaging LLM developers with bias mitigation.
Fichier principal
Vignette du fichier
Workshop_Amsterdam_0924.pdf (648.44 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-04755691 , version 1 (28-10-2024)

Licence

Identifiers

  • HAL Id : hal-04755691 , version 1

Cite

Fanny Ducel, Aurélie Névéol, Karën Fort. Desiderata for Actionable Bias Research. New Perspectives on Bias and Discrimination in Language Technology, Nov 2024, Amsterdam (Pays-Bas), France. ⟨hal-04755691⟩
0 View
0 Download

Share

More