<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="gao-etal-2024-insights">
    <titleInfo>
        <title>Insights into LLM Long-Context Failures: When Transformers Know but Don’t Tell</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Muhan</namePart>
        <namePart type="family">Gao</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">TaiMing</namePart>
        <namePart type="family">Lu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Kuai</namePart>
        <namePart type="family">Yu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Adam</namePart>
        <namePart type="family">Byerly</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Daniel</namePart>
        <namePart type="family">Khashabi</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2024-11</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Findings of the Association for Computational Linguistics: EMNLP 2024</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Yaser</namePart>
            <namePart type="family">Al-Onaizan</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Mohit</namePart>
            <namePart type="family">Bansal</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Yun-Nung</namePart>
            <namePart type="family">Chen</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Miami, Florida, USA</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>Large Language Models (LLMs) exhibit positional bias, struggling to utilize information from the middle or end of long contexts. Our study explores LLMs’ long-context reasoning by probing their hidden representations. We find that while LLMs encode the position of target information, they often fail to leverage this in generating accurate responses. This reveals a disconnect between information retrieval and utilization, a “know but don’t tell” phenomenon. We further analyze the relationship between extraction time and final accuracy, offering insights into the underlying mechanics of transformer models.</abstract>
    <identifier type="citekey">gao-etal-2024-insights</identifier>
    <location>
        <url>https://aclanthology.org/2024.findings-emnlp.447</url>
    </location>
    <part>
        <date>2024-11</date>
        <extent unit="page">
            <start>7611</start>
            <end>7625</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Insights into LLM Long-Context Failures: When Transformers Know but Don’t Tell
%A Gao, Muhan
%A Lu, TaiMing
%A Yu, Kuai
%A Byerly, Adam
%A Khashabi, Daniel
%Y Al-Onaizan, Yaser
%Y Bansal, Mohit
%Y Chen, Yun-Nung
%S Findings of the Association for Computational Linguistics: EMNLP 2024
%D 2024
%8 November
%I Association for Computational Linguistics
%C Miami, Florida, USA
%F gao-etal-2024-insights
%X Large Language Models (LLMs) exhibit positional bias, struggling to utilize information from the middle or end of long contexts. Our study explores LLMs’ long-context reasoning by probing their hidden representations. We find that while LLMs encode the position of target information, they often fail to leverage this in generating accurate responses. This reveals a disconnect between information retrieval and utilization, a “know but don’t tell” phenomenon. We further analyze the relationship between extraction time and final accuracy, offering insights into the underlying mechanics of transformer models.
%U https://aclanthology.org/2024.findings-emnlp.447
%P 7611-7625

Markdown (Informal)

[Insights into LLM Long-Context Failures: When Transformers Know but Don’t Tell](https://aclanthology.org/2024.findings-emnlp.447) (Gao et al., Findings 2024)

• Insights into LLM Long-Context Failures: When Transformers Know but Don’t Tell (Gao et al., Findings 2024)

ACL

• Muhan Gao, TaiMing Lu, Kuai Yu, Adam Byerly, and Daniel Khashabi. 2024. Insights into LLM Long-Context Failures: When Transformers Know but Don’t Tell. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 7611–7625, Miami, Florida, USA. Association for Computational Linguistics.