Saved in:
Bibliographic Details
Main Authors: Kantharuban, Anjali, Milbauer, Jeremiah, Sap, Maarten, Strubell, Emma, Neubig, Graham
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2410.05613
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912404468662272
author Kantharuban, Anjali
Milbauer, Jeremiah
Sap, Maarten
Strubell, Emma
Neubig, Graham
author_facet Kantharuban, Anjali
Milbauer, Jeremiah
Sap, Maarten
Strubell, Emma
Neubig, Graham
contents While personalized recommendations are often desired by users, it can be difficult in practice to distinguish cases of bias from cases of personalization: we find that models generate racially stereotypical recommendations regardless of whether the user revealed their identity intentionally through explicit indications or unintentionally through implicit cues. We demonstrate that when people use large language models (LLMs) to generate recommendations, the LLMs produce responses that reflect both what the user wants and who the user is. We argue that chatbots ought to transparently indicate when recommendations are influenced by a user's revealed identity characteristics, but observe that they currently fail to do so. Our experiments show that even though a user's revealed identity significantly influences model recommendations (p < 0.001), model responses obfuscate this fact in response to user queries. This bias and lack of transparency occurs consistently across multiple popular consumer LLMs and for four American racial groups.
format Preprint
id arxiv_https___arxiv_org_abs_2410_05613
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Stereotype or Personalization? User Identity Biases Chatbot Recommendations
Kantharuban, Anjali
Milbauer, Jeremiah
Sap, Maarten
Strubell, Emma
Neubig, Graham
Computation and Language
While personalized recommendations are often desired by users, it can be difficult in practice to distinguish cases of bias from cases of personalization: we find that models generate racially stereotypical recommendations regardless of whether the user revealed their identity intentionally through explicit indications or unintentionally through implicit cues. We demonstrate that when people use large language models (LLMs) to generate recommendations, the LLMs produce responses that reflect both what the user wants and who the user is. We argue that chatbots ought to transparently indicate when recommendations are influenced by a user's revealed identity characteristics, but observe that they currently fail to do so. Our experiments show that even though a user's revealed identity significantly influences model recommendations (p < 0.001), model responses obfuscate this fact in response to user queries. This bias and lack of transparency occurs consistently across multiple popular consumer LLMs and for four American racial groups.
title Stereotype or Personalization? User Identity Biases Chatbot Recommendations
topic Computation and Language
url https://arxiv.org/abs/2410.05613