<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ziegler-etal-2025-craft">
    <titleInfo>
        <title>CRAFT Your Dataset: Task-Specific Synthetic Dataset Generation Through Corpus Retrieval and Augmentation</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Ingo</namePart>
        <namePart type="family">Ziegler</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Abdullatif</namePart>
        <namePart type="family">Köksal</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Desmond</namePart>
        <namePart type="family">Elliott</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Hinrich</namePart>
        <namePart type="family">Schütze</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2025</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <genre authority="bibutilsgt">journal article</genre>
    <relatedItem type="host">
        <titleInfo>
            <title>Transactions of the Association for Computational Linguistics</title>
        </titleInfo>
        <originInfo>
            <issuance>continuing</issuance>
            <publisher>MIT Press</publisher>
            <place>
                <placeTerm type="text">Cambridge, MA</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">periodical</genre>
        <genre authority="bibutilsgt">academic journal</genre>
    </relatedItem>
    <abstract>Building high-quality datasets for specialized tasks is a time-consuming and resource-intensive process that often requires specialized domain knowledge. We propose Corpus Retrieval and Augmentation for Fine-Tuning (CRAFT), a method for generating synthetic datasets, given a small number of user-written few-shots that demonstrate the task to be performed. Given these examples, CRAFT uses large-scale public web-crawled corpora and similarity-based document retrieval to find other relevant human-written documents. Lastly, instruction-tuned large language models (LLMs) augment the retrieved documents into custom-formatted task samples, which then can be used for fine- tuning. We demonstrate that CRAFT can efficiently generate large-scale task-specific training datasets for four diverse tasks: biology, medicine, and commonsense question-answering (QA), as well as summarization. Our experiments show that CRAFT-based models outperform or match general LLMs on QA tasks, while exceeding models trained on human-curated summarization data by 46 preference points. CRAFT outperforms other synthetic dataset generation methods such as Self- and Evol-Instruct, and remains robust even when the quality of the initial few-shots varies.</abstract>
    <identifier type="citekey">ziegler-etal-2025-craft</identifier>
    <identifier type="doi">10.1162/tacl.a.56</identifier>
    <location>
        <url>https://aclanthology.org/2025.tacl-1.76/</url>
    </location>
    <part>
        <date>2025</date>
        <detail type="volume"><number>13</number></detail>
        <extent unit="page">
            <start>1693</start>
            <end>1721</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Journal Article
%T CRAFT Your Dataset: Task-Specific Synthetic Dataset Generation Through Corpus Retrieval and Augmentation
%A Ziegler, Ingo
%A Köksal, Abdullatif
%A Elliott, Desmond
%A Schütze, Hinrich
%J Transactions of the Association for Computational Linguistics
%D 2025
%V 13
%I MIT Press
%C Cambridge, MA
%F ziegler-etal-2025-craft
%X Building high-quality datasets for specialized tasks is a time-consuming and resource-intensive process that often requires specialized domain knowledge. We propose Corpus Retrieval and Augmentation for Fine-Tuning (CRAFT), a method for generating synthetic datasets, given a small number of user-written few-shots that demonstrate the task to be performed. Given these examples, CRAFT uses large-scale public web-crawled corpora and similarity-based document retrieval to find other relevant human-written documents. Lastly, instruction-tuned large language models (LLMs) augment the retrieved documents into custom-formatted task samples, which then can be used for fine- tuning. We demonstrate that CRAFT can efficiently generate large-scale task-specific training datasets for four diverse tasks: biology, medicine, and commonsense question-answering (QA), as well as summarization. Our experiments show that CRAFT-based models outperform or match general LLMs on QA tasks, while exceeding models trained on human-curated summarization data by 46 preference points. CRAFT outperforms other synthetic dataset generation methods such as Self- and Evol-Instruct, and remains robust even when the quality of the initial few-shots varies.
%R 10.1162/tacl.a.56
%U https://aclanthology.org/2025.tacl-1.76/
%U https://doi.org/10.1162/tacl.a.56
%P 1693-1721

Markdown (Informal)

[CRAFT Your Dataset: Task-Specific Synthetic Dataset Generation Through Corpus Retrieval and Augmentation](https://aclanthology.org/2025.tacl-1.76/) (Ziegler et al., TACL 2025)

• CRAFT Your Dataset: Task-Specific Synthetic Dataset Generation Through Corpus Retrieval and Augmentation (Ziegler et al., TACL 2025)

ACL

• Ingo Ziegler, Abdullatif Köksal, Desmond Elliott, and Hinrich Schütze. 2025. CRAFT Your Dataset: Task-Specific Synthetic Dataset Generation Through Corpus Retrieval and Augmentation. Transactions of the Association for Computational Linguistics, 13:1693–1721.