<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="chrysostomou-etal-2024-investigating">
    <titleInfo>
        <title>Investigating Hallucinations in Pruned Large Language Models for Abstractive Summarization</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">George</namePart>
        <namePart type="family">Chrysostomou</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Zhixue</namePart>
        <namePart type="family">Zhao</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Miles</namePart>
        <namePart type="family">Williams</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Nikolaos</namePart>
        <namePart type="family">Aletras</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2024</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <genre authority="bibutilsgt">journal article</genre>
    <relatedItem type="host">
        <titleInfo>
            <title>Transactions of the Association for Computational Linguistics</title>
        </titleInfo>
        <originInfo>
            <issuance>continuing</issuance>
            <publisher>MIT Press</publisher>
            <place>
                <placeTerm type="text">Cambridge, MA</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">periodical</genre>
        <genre authority="bibutilsgt">academic journal</genre>
    </relatedItem>
    <abstract>Despite the remarkable performance of generative large language models (LLMs) on abstractive summarization, they face two significant challenges: their considerable size and tendency to hallucinate. Hallucinations are concerning because they erode reliability and raise safety issues. Pruning is a technique that reduces model size by removing redundant weights, enabling more efficient sparse inference. Pruned models yield downstream task performance comparable to the original, making them ideal alternatives when operating on a limited budget. However, the effect that pruning has upon hallucinations in abstractive summarization with LLMs has yet to be explored. In this paper, we provide an extensive empirical study across five summarization datasets, two state-of-the-art pruning methods, and five instruction-tuned LLMs. Surprisingly, we find that hallucinations are less prevalent from pruned LLMs than the original models. Our analysis suggests that pruned models tend to depend more on the source document for summary generation. This leads to a higher lexical overlap between the generated summary and the source document, which could be a reason for the reduction in hallucination risk.1</abstract>
    <identifier type="citekey">chrysostomou-etal-2024-investigating</identifier>
    <identifier type="doi">10.1162/tacl_a_00695</identifier>
    <location>
        <url>https://aclanthology.org/2024.tacl-1.64</url>
    </location>
    <part>
        <date>2024</date>
        <detail type="volume"><number>12</number></detail>
        <extent unit="page">
            <start>1163</start>
            <end>1181</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Journal Article
%T Investigating Hallucinations in Pruned Large Language Models for Abstractive Summarization
%A Chrysostomou, George
%A Zhao, Zhixue
%A Williams, Miles
%A Aletras, Nikolaos
%J Transactions of the Association for Computational Linguistics
%D 2024
%V 12
%I MIT Press
%C Cambridge, MA
%F chrysostomou-etal-2024-investigating
%X Despite the remarkable performance of generative large language models (LLMs) on abstractive summarization, they face two significant challenges: their considerable size and tendency to hallucinate. Hallucinations are concerning because they erode reliability and raise safety issues. Pruning is a technique that reduces model size by removing redundant weights, enabling more efficient sparse inference. Pruned models yield downstream task performance comparable to the original, making them ideal alternatives when operating on a limited budget. However, the effect that pruning has upon hallucinations in abstractive summarization with LLMs has yet to be explored. In this paper, we provide an extensive empirical study across five summarization datasets, two state-of-the-art pruning methods, and five instruction-tuned LLMs. Surprisingly, we find that hallucinations are less prevalent from pruned LLMs than the original models. Our analysis suggests that pruned models tend to depend more on the source document for summary generation. This leads to a higher lexical overlap between the generated summary and the source document, which could be a reason for the reduction in hallucination risk.1
%R 10.1162/tacl_a_00695
%U https://aclanthology.org/2024.tacl-1.64
%U https://doi.org/10.1162/tacl_a_00695
%P 1163-1181

Markdown (Informal)

[Investigating Hallucinations in Pruned Large Language Models for Abstractive Summarization](https://aclanthology.org/2024.tacl-1.64) (Chrysostomou et al., TACL 2024)

• Investigating Hallucinations in Pruned Large Language Models for Abstractive Summarization (Chrysostomou et al., TACL 2024)

ACL

• George Chrysostomou, Zhixue Zhao, Miles Williams, and Nikolaos Aletras. 2024. Investigating Hallucinations in Pruned Large Language Models for Abstractive Summarization. Transactions of the Association for Computational Linguistics, 12:1163–1181.