<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zheng-shi-2026-relation">
    <titleInfo>
        <title>From Relation-Specific Heads to Shared Interventions: Contextual Entrainment in Language Models</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Jinghan</namePart>
        <namePart type="family">Zheng</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jiajun</namePart>
        <namePart type="family">Shi</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-11</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Yue</namePart>
            <namePart type="family">Zhang</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Ruifeng</namePart>
            <namePart type="family">Xu</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Ayu</namePart>
            <namePart type="family">Purwarianti</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Tong</namePart>
            <namePart type="family">Rong</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Jun</namePart>
            <namePart type="family">Suzuki</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Hengqin, China</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-439-2</identifier>
    </relatedItem>
    <abstract>Large language models rely on prompt context to perform in-context learning, follow formats, and adapt to local evidence. This ability, however, also raises a robustness concern: models may reuse irrelevant tokens from earlier context when answering a later query. Prior work identifies this behavior as contextual entrainment and finds attention heads whose ablation reduces it, arguing that these heads are largely task- or relation-specific. But are the mechanisms behind contextual entrainment truly isolated by relation, or do relation-conditioned searches recover reusable components shared across relations? Using factual-distractor relation-completion prompts across 19 lexical-semantic subtypes, we identify attention heads whose ablation reduces entrainment and test how the resulting head sets transfer recovery ability across exact relations, semantic families, and mixed relation distributions. Same-family transfer exceeds different-family transfer, family residuals weaken once globally shared heads are removed, and subtype-specific searches repeatedly rediscover many global heads. Clean-completion and context-use probes further suggest that this shared intervention can reduce harmful copying while preserving ordinary relation completion and simple in-context abilities. Our findings provide a more precise perspective on contextual entrainment and its mitigation: harmful context following is not best addressed by searching for isolated relation-specific circuits, but by using heterogeneous attack prompts to reveal reusable intervention targets whose effects are shaped by semantic relation structure.</abstract>
    <identifier type="citekey">zheng-shi-2026-relation</identifier>
    <location>
        <url>https://aclanthology.org/2026.aacl-long.149/</url>
    </location>
    <part>
        <date>2026-11</date>
        <extent unit="page">
            <start>3278</start>
            <end>3295</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T From Relation-Specific Heads to Shared Interventions: Contextual Entrainment in Language Models
%A Zheng, Jinghan
%A Shi, Jiajun
%Y Zhang, Yue
%Y Xu, Ruifeng
%Y Purwarianti, Ayu
%Y Rong, Tong
%Y Suzuki, Jun
%S Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2026
%8 November
%I Association for Computational Linguistics
%C Hengqin, China
%@ 979-8-89176-439-2
%F zheng-shi-2026-relation
%X Large language models rely on prompt context to perform in-context learning, follow formats, and adapt to local evidence. This ability, however, also raises a robustness concern: models may reuse irrelevant tokens from earlier context when answering a later query. Prior work identifies this behavior as contextual entrainment and finds attention heads whose ablation reduces it, arguing that these heads are largely task- or relation-specific. But are the mechanisms behind contextual entrainment truly isolated by relation, or do relation-conditioned searches recover reusable components shared across relations? Using factual-distractor relation-completion prompts across 19 lexical-semantic subtypes, we identify attention heads whose ablation reduces entrainment and test how the resulting head sets transfer recovery ability across exact relations, semantic families, and mixed relation distributions. Same-family transfer exceeds different-family transfer, family residuals weaken once globally shared heads are removed, and subtype-specific searches repeatedly rediscover many global heads. Clean-completion and context-use probes further suggest that this shared intervention can reduce harmful copying while preserving ordinary relation completion and simple in-context abilities. Our findings provide a more precise perspective on contextual entrainment and its mitigation: harmful context following is not best addressed by searching for isolated relation-specific circuits, but by using heterogeneous attack prompts to reveal reusable intervention targets whose effects are shaped by semantic relation structure.
%U https://aclanthology.org/2026.aacl-long.149/
%P 3278-3295

Markdown (Informal)

[From Relation-Specific Heads to Shared Interventions: Contextual Entrainment in Language Models](https://aclanthology.org/2026.aacl-long.149/) (Zheng & Shi, AACL-IJCNLP 2026)

• From Relation-Specific Heads to Shared Interventions: Contextual Entrainment in Language Models (Zheng & Shi, AACL-IJCNLP 2026)

ACL

• Jinghan Zheng and Jiajun Shi. 2026. From Relation-Specific Heads to Shared Interventions: Contextual Entrainment in Language Models. In Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers), pages 3278–3295, Hengqin, China. Association for Computational Linguistics.