<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="madsen-etal-2024-self">
    <titleInfo>
        <title>Are self-explanations from Large Language Models faithful?</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Andreas</namePart>
        <namePart type="family">Madsen</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Sarath</namePart>
        <namePart type="family">Chandar</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Siva</namePart>
        <namePart type="family">Reddy</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2024-08</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Findings of the Association for Computational Linguistics: ACL 2024</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Lun-Wei</namePart>
            <namePart type="family">Ku</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Andre</namePart>
            <namePart type="family">Martins</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Vivek</namePart>
            <namePart type="family">Srikumar</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Bangkok, Thailand</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>Instruction-tuned Large Language Models (LLMs) excel at many tasks and will even explain their reasoning, so-called self-explanations. However, convincing and wrong self-explanations can lead to unsupported confidence in LLMs, thus increasing risk. Therefore, it‘s important to measure if self-explanations truly reflect the model‘s behavior. Such a measure is called interpretability-faithfulness and is challenging to perform since the ground truth is inaccessible, and many LLMs only have an inference API. To address this, we propose employing self-consistency checks to measure faithfulness. For example, if an LLM says a set of words is important for making a prediction, then it should not be able to make its prediction without these words. While self-consistency checks are a common approach to faithfulness, they have not previously been successfully applied to LLM self-explanations for counterfactual, feature attribution, and redaction explanations. Our results demonstrate that faithfulness is explanation, model, and task-dependent, showing self-explanations should not be trusted in general. For example, with sentiment classification, counterfactuals are more faithful for Llama2, feature attribution for Mistral, and redaction for Falcon 40B.</abstract>
    <identifier type="citekey">madsen-etal-2024-self</identifier>
    <identifier type="doi">10.18653/v1/2024.findings-acl.19</identifier>
    <location>
        <url>https://aclanthology.org/2024.findings-acl.19/</url>
    </location>
    <part>
        <date>2024-08</date>
        <extent unit="page">
            <start>295</start>
            <end>337</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Are self-explanations from Large Language Models faithful?
%A Madsen, Andreas
%A Chandar, Sarath
%A Reddy, Siva
%Y Ku, Lun-Wei
%Y Martins, Andre
%Y Srikumar, Vivek
%S Findings of the Association for Computational Linguistics: ACL 2024
%D 2024
%8 August
%I Association for Computational Linguistics
%C Bangkok, Thailand
%F madsen-etal-2024-self
%X Instruction-tuned Large Language Models (LLMs) excel at many tasks and will even explain their reasoning, so-called self-explanations. However, convincing and wrong self-explanations can lead to unsupported confidence in LLMs, thus increasing risk. Therefore, it‘s important to measure if self-explanations truly reflect the model‘s behavior. Such a measure is called interpretability-faithfulness and is challenging to perform since the ground truth is inaccessible, and many LLMs only have an inference API. To address this, we propose employing self-consistency checks to measure faithfulness. For example, if an LLM says a set of words is important for making a prediction, then it should not be able to make its prediction without these words. While self-consistency checks are a common approach to faithfulness, they have not previously been successfully applied to LLM self-explanations for counterfactual, feature attribution, and redaction explanations. Our results demonstrate that faithfulness is explanation, model, and task-dependent, showing self-explanations should not be trusted in general. For example, with sentiment classification, counterfactuals are more faithful for Llama2, feature attribution for Mistral, and redaction for Falcon 40B.
%R 10.18653/v1/2024.findings-acl.19
%U https://aclanthology.org/2024.findings-acl.19/
%U https://doi.org/10.18653/v1/2024.findings-acl.19
%P 295-337

Markdown (Informal)

[Are self-explanations from Large Language Models faithful?](https://aclanthology.org/2024.findings-acl.19/) (Madsen et al., Findings 2024)

• Are self-explanations from Large Language Models faithful? (Madsen et al., Findings 2024)

ACL

• Andreas Madsen, Sarath Chandar, and Siva Reddy. 2024. Are self-explanations from Large Language Models faithful?. In Findings of the Association for Computational Linguistics: ACL 2024, pages 295–337, Bangkok, Thailand. Association for Computational Linguistics.