<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="huang-etal-2026-challenging">
    <titleInfo>
        <title>Challenging the Explanation Based on Preceding Tokens: Discovering Transferable Non-Literal Biasing</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Yuchen</namePart>
        <namePart type="family">Huang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Junpeng</namePart>
        <namePart type="family">Zhang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Quanshi</namePart>
        <namePart type="family">Zhang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-07</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Maria</namePart>
            <namePart type="family">Liakata</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Viviane</namePart>
            <namePart type="given">P</namePart>
            <namePart type="family">Moreira</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Jiajun</namePart>
            <namePart type="family">Zhang</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">David</namePart>
            <namePart type="family">Jurgens</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">San Diego, California, United States</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-391-3</identifier>
    </relatedItem>
    <abstract>In this paper, we find that the generated preceding tokens, which are not directly related to the answer, may still significantly push the large language model (LLM) towards the target answer. More crucially, the biased connotations of target answer in the preceding tokens can also transfer to other prompts. This finding suggests that the LLM may intentionally use the semantically unrelated tokens to help the generation of the target answer. Our finding offers a new perspective on understanding the long-range dependency phenomena in LLMs.</abstract>
    <identifier type="citekey">huang-etal-2026-challenging</identifier>
    <location>
        <url>https://aclanthology.org/2026.acl-short.52/</url>
    </location>
    <part>
        <date>2026-07</date>
        <extent unit="page">
            <start>637</start>
            <end>645</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Challenging the Explanation Based on Preceding Tokens: Discovering Transferable Non-Literal Biasing
%A Huang, Yuchen
%A Zhang, Junpeng
%A Zhang, Quanshi
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-391-3
%F huang-etal-2026-challenging
%X In this paper, we find that the generated preceding tokens, which are not directly related to the answer, may still significantly push the large language model (LLM) towards the target answer. More crucially, the biased connotations of target answer in the preceding tokens can also transfer to other prompts. This finding suggests that the LLM may intentionally use the semantically unrelated tokens to help the generation of the target answer. Our finding offers a new perspective on understanding the long-range dependency phenomena in LLMs.
%U https://aclanthology.org/2026.acl-short.52/
%P 637-645

Markdown (Informal)

[Challenging the Explanation Based on Preceding Tokens: Discovering Transferable Non-Literal Biasing](https://aclanthology.org/2026.acl-short.52/) (Huang et al., ACL 2026)

• Challenging the Explanation Based on Preceding Tokens: Discovering Transferable Non-Literal Biasing (Huang et al., ACL 2026)

ACL

• Yuchen Huang, Junpeng Zhang, and Quanshi Zhang. 2026. Challenging the Explanation Based on Preceding Tokens: Discovering Transferable Non-Literal Biasing. In Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 637–645, San Diego, California, United States. Association for Computational Linguistics.