<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="aljaafari-etal-2026-emergence">
    <titleInfo>
        <title>Emergence and Localisation of Semantic Role Circuits in LLMs</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Nura</namePart>
        <namePart type="family">Aljaafari</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Danilo</namePart>
        <namePart type="family">Carvalho</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Andre</namePart>
        <namePart type="family">Freitas</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>Findings of the Association for Computational Linguistics: ACL 2026</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">979-8-89176-395-1</identifier>
    </relatedItem>
    <abstract>Despite displaying semantic competence, large language models’ internal mechanisms that ground abstract semantic structure remain insufficiently characterised. To investigate whether and how LLMs develop causally functional representations of semantic roles, we introduce a causal-temporal methodology combining contrastive minimal pairs, edge-attribution circuit discovery, and training-time tracking. Our analysis reveals that LLMs encode semantic roles through highly localised circuits (89–92% attribution within 28 nodes) that emerge gradually via structural refinement rather than phase transitions. These circuits exhibit moderate cross-scale conservation (24–51% component overlap) alongside high spectral similarity, with larger models reusing similar components while rewiring connections. These findings suggest that LLMs form compact, causally isolated mechanisms for abstract semantic structure that exhibit partial transfer across scales and architectures.</abstract>
    <identifier type="citekey">aljaafari-etal-2026-emergence</identifier>
    <identifier type="doi">10.18653/v1/2026.findings-acl.1964</identifier>
    <location>
        <url>https://aclanthology.org/2026.findings-acl.1964/</url>
    </location>
    <part>
        <date>2026-07</date>
        <extent unit="page">
            <start>39402</start>
            <end>39433</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Emergence and Localisation of Semantic Role Circuits in LLMs
%A Aljaafari, Nura
%A Carvalho, Danilo
%A Freitas, Andre
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Findings of the Association for Computational Linguistics: ACL 2026
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-395-1
%F aljaafari-etal-2026-emergence
%X Despite displaying semantic competence, large language models’ internal mechanisms that ground abstract semantic structure remain insufficiently characterised. To investigate whether and how LLMs develop causally functional representations of semantic roles, we introduce a causal-temporal methodology combining contrastive minimal pairs, edge-attribution circuit discovery, and training-time tracking. Our analysis reveals that LLMs encode semantic roles through highly localised circuits (89–92% attribution within 28 nodes) that emerge gradually via structural refinement rather than phase transitions. These circuits exhibit moderate cross-scale conservation (24–51% component overlap) alongside high spectral similarity, with larger models reusing similar components while rewiring connections. These findings suggest that LLMs form compact, causally isolated mechanisms for abstract semantic structure that exhibit partial transfer across scales and architectures.
%R 10.18653/v1/2026.findings-acl.1964
%U https://aclanthology.org/2026.findings-acl.1964/
%U https://doi.org/10.18653/v1/2026.findings-acl.1964
%P 39402-39433

Markdown (Informal)

[Emergence and Localisation of Semantic Role Circuits in LLMs](https://aclanthology.org/2026.findings-acl.1964/) (Aljaafari et al., Findings 2026)

• Emergence and Localisation of Semantic Role Circuits in LLMs (Aljaafari et al., Findings 2026)

ACL

• Nura Aljaafari, Danilo Carvalho, and Andre Freitas. 2026. Emergence and Localisation of Semantic Role Circuits in LLMs. In Findings of the Association for Computational Linguistics: ACL 2026, pages 39402–39433, San Diego, California, United States. Association for Computational Linguistics.