<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="dymkiewicz-etal-2026-donors">
    <titleInfo>
        <title>Donors and Recipients: On Asymmetric Transfer Across Tasks and Languages with Parameter-Efficient Fine-Tuning</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Kajetan</namePart>
        <namePart type="family">Dymkiewicz</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Ivan</namePart>
        <namePart type="family">Vulić</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Helen</namePart>
        <namePart type="family">Yannakoudakis</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Eilam</namePart>
        <namePart type="family">Shapira</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Roi</namePart>
        <namePart type="family">Reichart</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Anna</namePart>
        <namePart type="family">Korhonen</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>Findings of the Association for Computational Linguistics: EMNLP 2026</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">979-8-89176-462-0</identifier>
    </relatedItem>
    <abstract>Large language models (LLMs) perform strongly across tasks and languages, yet how improvements in one task or language affect other tasks and languages remains poorly understood. We conduct a controlled LoRA fine-tuning study across multiple open-weight LLM families and scales, using a standardised grid of 11 languages and four benchmarks. We fine-tune each model on a single task–language source, then evaluate it on all other task–language target pairs to measure transfer. We decompose transfer into three regimes: (i) Matched-Task (Cross-Language), (ii) Cross-Task (Matched-Language), and (iii) Cross-Task (Cross-Language). Single-source fine-tuning yields a net positive uplift across regimes, but the gains are strongly asymmetric. Matched-Task (Cross-Language) transfer emerges as the most effective and structurally regular regime, with transfer magnitude driven principally by the identity of the target language rather than model architecture. We identify a stable coarse-grained hierarchy in which some task and language targets consistently absorb gains from diverse sources, while others remain relatively isolated. These results imply that effective fine-tuning requires accounting for donor–recipient roles to maximise downstream gains while limiting collateral degradation in other capabilities.</abstract>
    <identifier type="citekey">dymkiewicz-etal-2026-donors</identifier>
    <location>
        <url>https://aclanthology.org/2026.findings-emnlp.67/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>1356</start>
            <end>1373</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Donors and Recipients: On Asymmetric Transfer Across Tasks and Languages with Parameter-Efficient Fine-Tuning
%A Dymkiewicz, Kajetan
%A Vulić, Ivan
%A Yannakoudakis, Helen
%A Shapira, Eilam
%A Reichart, Roi
%A Korhonen, Anna
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F dymkiewicz-etal-2026-donors
%X Large language models (LLMs) perform strongly across tasks and languages, yet how improvements in one task or language affect other tasks and languages remains poorly understood. We conduct a controlled LoRA fine-tuning study across multiple open-weight LLM families and scales, using a standardised grid of 11 languages and four benchmarks. We fine-tune each model on a single task–language source, then evaluate it on all other task–language target pairs to measure transfer. We decompose transfer into three regimes: (i) Matched-Task (Cross-Language), (ii) Cross-Task (Matched-Language), and (iii) Cross-Task (Cross-Language). Single-source fine-tuning yields a net positive uplift across regimes, but the gains are strongly asymmetric. Matched-Task (Cross-Language) transfer emerges as the most effective and structurally regular regime, with transfer magnitude driven principally by the identity of the target language rather than model architecture. We identify a stable coarse-grained hierarchy in which some task and language targets consistently absorb gains from diverse sources, while others remain relatively isolated. These results imply that effective fine-tuning requires accounting for donor–recipient roles to maximise downstream gains while limiting collateral degradation in other capabilities.
%U https://aclanthology.org/2026.findings-emnlp.67/
%P 1356-1373

Markdown (Informal)

[Donors and Recipients: On Asymmetric Transfer Across Tasks and Languages with Parameter-Efficient Fine-Tuning](https://aclanthology.org/2026.findings-emnlp.67/) (Dymkiewicz et al., Findings 2026)

• Donors and Recipients: On Asymmetric Transfer Across Tasks and Languages with Parameter-Efficient Fine-Tuning (Dymkiewicz et al., Findings 2026)

ACL

• Kajetan Dymkiewicz, Ivan Vulić, Helen Yannakoudakis, Eilam Shapira, Roi Reichart, and Anna Korhonen. 2026. Donors and Recipients: On Asymmetric Transfer Across Tasks and Languages with Parameter-Efficient Fine-Tuning. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 1356–1373, Budapest, Hungary. Association for Computational Linguistics.