<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="lu-etal-2025-paired">
    <titleInfo>
        <title>Paired by the Teacher: Turning Unpaired Data into High-Fidelity Pairs for Low-Resource Text Generation</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Yen-Ju</namePart>
        <namePart type="family">Lu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Thomas</namePart>
        <namePart type="family">Thebaud</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Laureano</namePart>
        <namePart type="family">Moro-Velazquez</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Najim</namePart>
        <namePart type="family">Dehak</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jesus</namePart>
        <namePart type="family">Villalba</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2025-11</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Christos</namePart>
            <namePart type="family">Christodoulopoulos</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Tanmoy</namePart>
            <namePart type="family">Chakraborty</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Carolyn</namePart>
            <namePart type="family">Rose</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Violet</namePart>
            <namePart type="family">Peng</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Suzhou, China</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-332-6</identifier>
    </relatedItem>
    <abstract>We present Paired by the Teacher (PbT), a two-stage teacher–student pipeline that synthesizes accurate input–output pairs without human labels or parallel data. In many low-resource natural language generation (NLG) scenarios, practitioners may have only raw outputs, like highlights, recaps, or questions, or only raw inputs, such as articles, dialogues, or paragraphs, but seldom both. This mismatch forces small models to learn from very few examples or rely on costly, broad-scope synthetic examples produced by large LLMs. PbT addresses this by asking a teacher LLM to compress each unpaired example into a concise intermediate representation (IR), and training a student to reconstruct inputs from IRs. This enables outputs to be paired with student-generated inputs, yielding high-quality synthetic data. We evaluate PbT on five benchmarks—document summarization (XSum, CNNDM), dialogue summarization (SAMSum, DialogSum), and question generation (SQuAD)—as well as an unpaired setting on SwitchBoard (paired with DialogSum summaries). An 8B student trained only on PbT data outperforms models trained on 70 B teacher-generated corpora and other unsupervised baselines, coming within 1.2 ROUGE-L of human-annotated pairs and closing 82% of the oracle gap at one-third the annotation cost of direct synthesis. Human evaluation on SwitchBoard further confirms that only PbT produces concise, faithful summaries aligned with the target style, highlighting its advantage of generating in-domain sources that avoid the mismatch, limiting direct synthesis.</abstract>
    <identifier type="citekey">lu-etal-2025-paired</identifier>
    <location>
        <url>https://aclanthology.org/2025.emnlp-main.1030/</url>
    </location>
    <part>
        <date>2025-11</date>
        <extent unit="page">
            <start>20412</start>
            <end>20434</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Paired by the Teacher: Turning Unpaired Data into High-Fidelity Pairs for Low-Resource Text Generation
%A Lu, Yen-Ju
%A Thebaud, Thomas
%A Moro-Velazquez, Laureano
%A Dehak, Najim
%A Villalba, Jesus
%Y Christodoulopoulos, Christos
%Y Chakraborty, Tanmoy
%Y Rose, Carolyn
%Y Peng, Violet
%S Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou, China
%@ 979-8-89176-332-6
%F lu-etal-2025-paired
%X We present Paired by the Teacher (PbT), a two-stage teacher–student pipeline that synthesizes accurate input–output pairs without human labels or parallel data. In many low-resource natural language generation (NLG) scenarios, practitioners may have only raw outputs, like highlights, recaps, or questions, or only raw inputs, such as articles, dialogues, or paragraphs, but seldom both. This mismatch forces small models to learn from very few examples or rely on costly, broad-scope synthetic examples produced by large LLMs. PbT addresses this by asking a teacher LLM to compress each unpaired example into a concise intermediate representation (IR), and training a student to reconstruct inputs from IRs. This enables outputs to be paired with student-generated inputs, yielding high-quality synthetic data. We evaluate PbT on five benchmarks—document summarization (XSum, CNNDM), dialogue summarization (SAMSum, DialogSum), and question generation (SQuAD)—as well as an unpaired setting on SwitchBoard (paired with DialogSum summaries). An 8B student trained only on PbT data outperforms models trained on 70 B teacher-generated corpora and other unsupervised baselines, coming within 1.2 ROUGE-L of human-annotated pairs and closing 82% of the oracle gap at one-third the annotation cost of direct synthesis. Human evaluation on SwitchBoard further confirms that only PbT produces concise, faithful summaries aligned with the target style, highlighting its advantage of generating in-domain sources that avoid the mismatch, limiting direct synthesis.
%U https://aclanthology.org/2025.emnlp-main.1030/
%P 20412-20434

Markdown (Informal)

[Paired by the Teacher: Turning Unpaired Data into High-Fidelity Pairs for Low-Resource Text Generation](https://aclanthology.org/2025.emnlp-main.1030/) (Lu et al., EMNLP 2025)

• Paired by the Teacher: Turning Unpaired Data into High-Fidelity Pairs for Low-Resource Text Generation (Lu et al., EMNLP 2025)

ACL

• Yen-Ju Lu, Thomas Thebaud, Laureano Moro-Velazquez, Najim Dehak, and Jesus Villalba. 2025. Paired by the Teacher: Turning Unpaired Data into High-Fidelity Pairs for Low-Resource Text Generation. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 20412–20434, Suzhou, China. Association for Computational Linguistics.