<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wang-etal-2025-position">
    <titleInfo>
        <title>POSITION BIAS MITIGATES POSITION BIAS: Mitigate Position Bias Through Inter-Position Knowledge Distillation</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Yifei</namePart>
        <namePart type="family">Wang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Feng</namePart>
        <namePart type="family">Xiong</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yong</namePart>
        <namePart type="family">Wang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Linjing</namePart>
        <namePart type="family">Li</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Xiangxiang</namePart>
        <namePart type="family">Chu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Daniel</namePart>
        <namePart type="given">Dajun</namePart>
        <namePart type="family">Zeng</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2025-11</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Christos</namePart>
            <namePart type="family">Christodoulopoulos</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Tanmoy</namePart>
            <namePart type="family">Chakraborty</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Carolyn</namePart>
            <namePart type="family">Rose</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Violet</namePart>
            <namePart type="family">Peng</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Suzhou, China</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-332-6</identifier>
    </relatedItem>
    <abstract>Positional bias (PB), manifesting as non-uniform sensitivity across different contextual locations, significantly impairs long-context comprehension and processing capabilities. Previous studies have addressed PB either by modifying the underlying architectures or by employing extensive contextual awareness training. However, the former approach fails to effectively eliminate the substantialperformance disparities, while the latter imposes significant data and computational overhead. To address PB effectively, we introduce Pos2Distill, a position to position knowledge distillation framework. Pos2Distill transfers the superior capabilities from advantageous positions to less favorable ones, thereby reducing the huge performance gaps. The conceptual principle is to leverage the inherent, position-induced disparity to counteract the PB itself. We identify distinct manifestations of PB under retrieval and reasoning paradigms, thereby designing two specialized instantiations: Pos2Distill-R1 and Pos2Distill-R2 respectively, both grounded in this core principle. By employing the Pos2Distill approach, we achieve enhanced uniformity and significant performance gains across all contextual positions in long-context retrieval and reasoning tasks. Crucially, both specialized systems exhibit strong cross-task generalization mutually, while achieving superior performance on their respective tasks.</abstract>
    <identifier type="citekey">wang-etal-2025-position</identifier>
    <location>
        <url>https://aclanthology.org/2025.emnlp-main.78/</url>
    </location>
    <part>
        <date>2025-11</date>
        <extent unit="page">
            <start>1495</start>
            <end>1512</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T POSITION BIAS MITIGATES POSITION BIAS: Mitigate Position Bias Through Inter-Position Knowledge Distillation
%A Wang, Yifei
%A Xiong, Feng
%A Wang, Yong
%A Li, Linjing
%A Chu, Xiangxiang
%A Zeng, Daniel Dajun
%Y Christodoulopoulos, Christos
%Y Chakraborty, Tanmoy
%Y Rose, Carolyn
%Y Peng, Violet
%S Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou, China
%@ 979-8-89176-332-6
%F wang-etal-2025-position
%X Positional bias (PB), manifesting as non-uniform sensitivity across different contextual locations, significantly impairs long-context comprehension and processing capabilities. Previous studies have addressed PB either by modifying the underlying architectures or by employing extensive contextual awareness training. However, the former approach fails to effectively eliminate the substantialperformance disparities, while the latter imposes significant data and computational overhead. To address PB effectively, we introduce Pos2Distill, a position to position knowledge distillation framework. Pos2Distill transfers the superior capabilities from advantageous positions to less favorable ones, thereby reducing the huge performance gaps. The conceptual principle is to leverage the inherent, position-induced disparity to counteract the PB itself. We identify distinct manifestations of PB under retrieval and reasoning paradigms, thereby designing two specialized instantiations: Pos2Distill-R1 and Pos2Distill-R2 respectively, both grounded in this core principle. By employing the Pos2Distill approach, we achieve enhanced uniformity and significant performance gains across all contextual positions in long-context retrieval and reasoning tasks. Crucially, both specialized systems exhibit strong cross-task generalization mutually, while achieving superior performance on their respective tasks.
%U https://aclanthology.org/2025.emnlp-main.78/
%P 1495-1512

Markdown (Informal)

[POSITION BIAS MITIGATES POSITION BIAS: Mitigate Position Bias Through Inter-Position Knowledge Distillation](https://aclanthology.org/2025.emnlp-main.78/) (Wang et al., EMNLP 2025)

• POSITION BIAS MITIGATES POSITION BIAS: Mitigate Position Bias Through Inter-Position Knowledge Distillation (Wang et al., EMNLP 2025)

ACL

• Yifei Wang, Feng Xiong, Yong Wang, Linjing Li, Xiangxiang Chu, and Daniel Dajun Zeng. 2025. POSITION BIAS MITIGATES POSITION BIAS: Mitigate Position Bias Through Inter-Position Knowledge Distillation. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 1495–1512, Suzhou, China. Association for Computational Linguistics.