@inproceedings{bauer-etal-2024-believe,
title = "{BELIEVE}: Belief-Enhanced Instruction Generation and Augmentation for Zero-Shot Bias Mitigation",
author = "Bauer, Lisa and
Mehrabi, Ninareh and
Goyal, Palash and
Chang, Kai-Wei and
Galstyan, Aram and
Gupta, Rahul",
editor = "Ovalle, Anaelia and
Chang, Kai-Wei and
Cao, Yang Trista and
Mehrabi, Ninareh and
Zhao, Jieyu and
Galstyan, Aram and
Dhamala, Jwala and
Kumar, Anoop and
Gupta, Rahul",
booktitle = "Proceedings of the 4th Workshop on Trustworthy Natural Language Processing (TrustNLP 2024)",
month = jun,
year = "2024",
address = "Mexico City, Mexico",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.trustnlp-1.20",
doi = "10.18653/v1/2024.trustnlp-1.20",
pages = "239--251",
abstract = "Language models, pre-trained on large amounts of unmoderated content, have been shown to contain societal biases. Mitigating such biases typically requires access to model parameters and training schemas. In this work, we address bias mitigation at inference time, such that it can be applied to any black-box model. To this end, we propose a belief generation and augmentation framework, BELIEVE, that demonstrates effective bias mitigation for natural language generation by augmenting input prompts with automatically generated instruction-based beliefs. Our framework eases the bottleneck required for manually crafting these instruction-based beliefs, by extending a recently proposed iterative in-context learning framework to automatically generate beliefs via a language model. We assess the impact of this system on fairness, and demonstrate effective bias mitigation on pretrained and instruction-tuned models for both sentiment and regard with respect to multiple protected classes including race, gender, and political ideology.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="bauer-etal-2024-believe">
<titleInfo>
<title>BELIEVE: Belief-Enhanced Instruction Generation and Augmentation for Zero-Shot Bias Mitigation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Lisa</namePart>
<namePart type="family">Bauer</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ninareh</namePart>
<namePart type="family">Mehrabi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Palash</namePart>
<namePart type="family">Goyal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kai-Wei</namePart>
<namePart type="family">Chang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aram</namePart>
<namePart type="family">Galstyan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rahul</namePart>
<namePart type="family">Gupta</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-06</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 4th Workshop on Trustworthy Natural Language Processing (TrustNLP 2024)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Anaelia</namePart>
<namePart type="family">Ovalle</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kai-Wei</namePart>
<namePart type="family">Chang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yang</namePart>
<namePart type="given">Trista</namePart>
<namePart type="family">Cao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ninareh</namePart>
<namePart type="family">Mehrabi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jieyu</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aram</namePart>
<namePart type="family">Galstyan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jwala</namePart>
<namePart type="family">Dhamala</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Anoop</namePart>
<namePart type="family">Kumar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rahul</namePart>
<namePart type="family">Gupta</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Mexico City, Mexico</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Language models, pre-trained on large amounts of unmoderated content, have been shown to contain societal biases. Mitigating such biases typically requires access to model parameters and training schemas. In this work, we address bias mitigation at inference time, such that it can be applied to any black-box model. To this end, we propose a belief generation and augmentation framework, BELIEVE, that demonstrates effective bias mitigation for natural language generation by augmenting input prompts with automatically generated instruction-based beliefs. Our framework eases the bottleneck required for manually crafting these instruction-based beliefs, by extending a recently proposed iterative in-context learning framework to automatically generate beliefs via a language model. We assess the impact of this system on fairness, and demonstrate effective bias mitigation on pretrained and instruction-tuned models for both sentiment and regard with respect to multiple protected classes including race, gender, and political ideology.</abstract>
<identifier type="citekey">bauer-etal-2024-believe</identifier>
<identifier type="doi">10.18653/v1/2024.trustnlp-1.20</identifier>
<location>
<url>https://aclanthology.org/2024.trustnlp-1.20</url>
</location>
<part>
<date>2024-06</date>
<extent unit="page">
<start>239</start>
<end>251</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T BELIEVE: Belief-Enhanced Instruction Generation and Augmentation for Zero-Shot Bias Mitigation
%A Bauer, Lisa
%A Mehrabi, Ninareh
%A Goyal, Palash
%A Chang, Kai-Wei
%A Galstyan, Aram
%A Gupta, Rahul
%Y Ovalle, Anaelia
%Y Chang, Kai-Wei
%Y Cao, Yang Trista
%Y Mehrabi, Ninareh
%Y Zhao, Jieyu
%Y Galstyan, Aram
%Y Dhamala, Jwala
%Y Kumar, Anoop
%Y Gupta, Rahul
%S Proceedings of the 4th Workshop on Trustworthy Natural Language Processing (TrustNLP 2024)
%D 2024
%8 June
%I Association for Computational Linguistics
%C Mexico City, Mexico
%F bauer-etal-2024-believe
%X Language models, pre-trained on large amounts of unmoderated content, have been shown to contain societal biases. Mitigating such biases typically requires access to model parameters and training schemas. In this work, we address bias mitigation at inference time, such that it can be applied to any black-box model. To this end, we propose a belief generation and augmentation framework, BELIEVE, that demonstrates effective bias mitigation for natural language generation by augmenting input prompts with automatically generated instruction-based beliefs. Our framework eases the bottleneck required for manually crafting these instruction-based beliefs, by extending a recently proposed iterative in-context learning framework to automatically generate beliefs via a language model. We assess the impact of this system on fairness, and demonstrate effective bias mitigation on pretrained and instruction-tuned models for both sentiment and regard with respect to multiple protected classes including race, gender, and political ideology.
%R 10.18653/v1/2024.trustnlp-1.20
%U https://aclanthology.org/2024.trustnlp-1.20
%U https://doi.org/10.18653/v1/2024.trustnlp-1.20
%P 239-251
Markdown (Informal)
[BELIEVE: Belief-Enhanced Instruction Generation and Augmentation for Zero-Shot Bias Mitigation](https://aclanthology.org/2024.trustnlp-1.20) (Bauer et al., TrustNLP-WS 2024)
ACL