<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ko-etal-2026-harmolora">
    <titleInfo>
        <title>HarmoLoRA: Commit-Stage Update Admission in Low-Rank Adaptation for Continual Speech Recognition</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Ye-Eun</namePart>
        <namePart type="family">Ko</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Ji-Eun</namePart>
        <namePart type="family">Choi</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Dong-Hyun</namePart>
        <namePart type="family">Kim</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Joon-Hyuk</namePart>
        <namePart type="family">Chang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-10</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Hung-Yi</namePart>
            <namePart type="family">Lee</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Sunipa</namePart>
            <namePart type="family">Dev</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Jessy</namePart>
            <namePart type="family">Li</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Desmond</namePart>
            <namePart type="family">Elliott</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Budapest, Hungary</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-459-0</identifier>
    </relatedItem>
    <abstract>Replay-free continual automatic speech recognition (ASR) requires adapting to new speech domains while maintaining a single persistent model without old data, task identifiers, or inference-time routing. We identify commit-stage update admission as a distinct post-training problem for continual low-rank adaptation (LoRA): after a task-wise LoRA update has been trained, the system must decide which part of that update should become persistent in the deployed model. This decision is especially important in ASR, where domains share phonetic and transcription structure but still differ in speakers, channels, noise, speaking style, and lexical coverage. We propose HarmoLoRA, a rank-constrained commit operator that decomposes the trained current-task update with respect to previously committed directions, constructs a rank-closed residual commit path, and uses a lexicographic constrained rule to choose the smallest residual admission coefficient among near-best realized candidates. The selected update is committed to one merged ASR checkpoint, requiring no replay, task identifiers, routing, gating, or multi-branch selection at inference. Across the reported ASR orders, HarmoLoRA improves recognition performance and reduces forgetting relative to reproduced replay-free and LoRA-based continual adaptation baselines.</abstract>
    <identifier type="citekey">ko-etal-2026-harmolora</identifier>
    <location>
        <url>https://aclanthology.org/2026.emnlp-main.2631/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>59824</start>
            <end>59841</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T HarmoLoRA: Commit-Stage Update Admission in Low-Rank Adaptation for Continual Speech Recognition
%A Ko, Ye-Eun
%A Choi, Ji-Eun
%A Kim, Dong-Hyun
%A Chang, Joon-Hyuk
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F ko-etal-2026-harmolora
%X Replay-free continual automatic speech recognition (ASR) requires adapting to new speech domains while maintaining a single persistent model without old data, task identifiers, or inference-time routing. We identify commit-stage update admission as a distinct post-training problem for continual low-rank adaptation (LoRA): after a task-wise LoRA update has been trained, the system must decide which part of that update should become persistent in the deployed model. This decision is especially important in ASR, where domains share phonetic and transcription structure but still differ in speakers, channels, noise, speaking style, and lexical coverage. We propose HarmoLoRA, a rank-constrained commit operator that decomposes the trained current-task update with respect to previously committed directions, constructs a rank-closed residual commit path, and uses a lexicographic constrained rule to choose the smallest residual admission coefficient among near-best realized candidates. The selected update is committed to one merged ASR checkpoint, requiring no replay, task identifiers, routing, gating, or multi-branch selection at inference. Across the reported ASR orders, HarmoLoRA improves recognition performance and reduces forgetting relative to reproduced replay-free and LoRA-based continual adaptation baselines.
%U https://aclanthology.org/2026.emnlp-main.2631/
%P 59824-59841

Markdown (Informal)

[HarmoLoRA: Commit-Stage Update Admission in Low-Rank Adaptation for Continual Speech Recognition](https://aclanthology.org/2026.emnlp-main.2631/) (Ko et al., EMNLP 2026)

• HarmoLoRA: Commit-Stage Update Admission in Low-Rank Adaptation for Continual Speech Recognition (Ko et al., EMNLP 2026)

ACL

• Ye-Eun Ko, Ji-Eun Choi, Dong-Hyun Kim, and Joon-Hyuk Chang. 2026. HarmoLoRA: Commit-Stage Update Admission in Low-Rank Adaptation for Continual Speech Recognition. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 59824–59841, Budapest, Hungary. Association for Computational Linguistics.