<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="jung-jung-2026-tracing">
    <titleInfo>
        <title>Tracing Logit Trajectories Across Layer Depth: Dataset-Level Explainability for Language Models</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Jeesu</namePart>
        <namePart type="family">Jung</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Sangkeun</namePart>
        <namePart type="family">Jung</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 1: Long 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">979-8-89176-390-6</identifier>
    </relatedItem>
    <abstract>Sentence-level explanations can miss the bigger picture of how a black-box model behaves across data, which matters most for complex criteria like safety that cannot be defined by a single rule. We trace **Logit-Trajectory**, which tracks adjacent-layer logit updates as vectors and aggregates them into a reproducible dataset-level trajectory pattern, enabling depth-wise explainability through signals such as coherence and angular rotation. Across 6 languages and 5 NLP tasks, we show these trajectory summaries reveal consistent depth-wise patterns that divergence- and similarity-based baselines often wash out due to scalarization. As a case study where dataset-level intermediate decision structure matters, we evaluate safety classification, reporting both trajectory-level visual separability and classification performance.</abstract>
    <identifier type="citekey">jung-jung-2026-tracing</identifier>
    <location>
        <url>https://aclanthology.org/2026.acl-long.809/</url>
    </location>
    <part>
        <date>2026-07</date>
        <extent unit="page">
            <start>17800</start>
            <end>17823</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Tracing Logit Trajectories Across Layer Depth: Dataset-Level Explainability for Language Models
%A Jung, Jeesu
%A Jung, Sangkeun
%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 1: Long Papers)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-390-6
%F jung-jung-2026-tracing
%X Sentence-level explanations can miss the bigger picture of how a black-box model behaves across data, which matters most for complex criteria like safety that cannot be defined by a single rule. We trace **Logit-Trajectory**, which tracks adjacent-layer logit updates as vectors and aggregates them into a reproducible dataset-level trajectory pattern, enabling depth-wise explainability through signals such as coherence and angular rotation. Across 6 languages and 5 NLP tasks, we show these trajectory summaries reveal consistent depth-wise patterns that divergence- and similarity-based baselines often wash out due to scalarization. As a case study where dataset-level intermediate decision structure matters, we evaluate safety classification, reporting both trajectory-level visual separability and classification performance.
%U https://aclanthology.org/2026.acl-long.809/
%P 17800-17823

Markdown (Informal)

[Tracing Logit Trajectories Across Layer Depth: Dataset-Level Explainability for Language Models](https://aclanthology.org/2026.acl-long.809/) (Jung & Jung, ACL 2026)

• Tracing Logit Trajectories Across Layer Depth: Dataset-Level Explainability for Language Models (Jung & Jung, ACL 2026)

ACL

• Jeesu Jung and Sangkeun Jung. 2026. Tracing Logit Trajectories Across Layer Depth: Dataset-Level Explainability for Language Models. In Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 17800–17823, San Diego, California, United States. Association for Computational Linguistics.