<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="popovic-farber-2026-tracing">
    <titleInfo>
        <title>Tracing Relational Knowledge Recall in Large Language Models</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Nicholas</namePart>
        <namePart type="family">Popovič</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Michael</namePart>
        <namePart type="family">Färber</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>We study how large language models recall relational knowledge during text generation, with a focus on identifying latent representations suitable for relation classification via linear probes.Prior work shows how attention heads and MLPs interact to resolve subject, predicate, and object, but it remains unclear which representations support faithful linear relation classification and why some relation types are easier to capture linearly than others.We systematically evaluate different latent representations derived from attention head and MLP contributions, showing that per-head attention contributions to the residual stream are comparatively strong features for linear relation classification.Feature attribution analyses of the trained probes, as well as characteristics of the different relation types, reveal clear correlations between probe accuracy and relation specificity, entity connectedness, and how distributed the signal on which the probe relies is across attention heads.Finally, we show how token-level feature attribution of probe predictions can be used to reveal probe behavior in further detail.</abstract>
    <identifier type="citekey">popovic-farber-2026-tracing</identifier>
    <identifier type="doi">10.18653/v1/2026.findings-acl.2160</identifier>
    <location>
        <url>https://aclanthology.org/2026.findings-acl.2160/</url>
    </location>
    <part>
        <date>2026-07</date>
        <extent unit="page">
            <start>43490</start>
            <end>43509</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Tracing Relational Knowledge Recall in Large Language Models
%A Popovič, Nicholas
%A Färber, Michael
%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 popovic-farber-2026-tracing
%X We study how large language models recall relational knowledge during text generation, with a focus on identifying latent representations suitable for relation classification via linear probes.Prior work shows how attention heads and MLPs interact to resolve subject, predicate, and object, but it remains unclear which representations support faithful linear relation classification and why some relation types are easier to capture linearly than others.We systematically evaluate different latent representations derived from attention head and MLP contributions, showing that per-head attention contributions to the residual stream are comparatively strong features for linear relation classification.Feature attribution analyses of the trained probes, as well as characteristics of the different relation types, reveal clear correlations between probe accuracy and relation specificity, entity connectedness, and how distributed the signal on which the probe relies is across attention heads.Finally, we show how token-level feature attribution of probe predictions can be used to reveal probe behavior in further detail.
%R 10.18653/v1/2026.findings-acl.2160
%U https://aclanthology.org/2026.findings-acl.2160/
%U https://doi.org/10.18653/v1/2026.findings-acl.2160
%P 43490-43509

Markdown (Informal)

[Tracing Relational Knowledge Recall in Large Language Models](https://aclanthology.org/2026.findings-acl.2160/) (Popovič & Färber, Findings 2026)

• Tracing Relational Knowledge Recall in Large Language Models (Popovič & Färber, Findings 2026)

ACL

• Nicholas Popovič and Michael Färber. 2026. Tracing Relational Knowledge Recall in Large Language Models. In Findings of the Association for Computational Linguistics: ACL 2026, pages 43490–43509, San Diego, California, United States. Association for Computational Linguistics.