<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="sahak-etal-2023-state">
    <titleInfo>
        <title>A State-Vector Framework for Dataset Effects</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Esmat</namePart>
        <namePart type="family">Sahak</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Zining</namePart>
        <namePart type="family">Zhu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Frank</namePart>
        <namePart type="family">Rudzicz</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2023-12</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Houda</namePart>
            <namePart type="family">Bouamor</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Juan</namePart>
            <namePart type="family">Pino</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Kalika</namePart>
            <namePart type="family">Bali</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Singapore</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>The impressive success of recent deep neural network (DNN)-based systems is significantly influenced by the high-quality datasets used in training. However, the effects of the datasets, especially how they interact with each other, remain underexplored. We propose a state-vector framework to enable rigorous studies in this direction. This framework uses idealized probing test results as the bases of a vector space. This framework allows us to quantify the effects of both standalone and interacting datasets. We show that the significant effects of some commonly-used language understanding datasets are characteristic and are concentrated on a few linguistic dimensions. Additionally, we observe some “spill-over” effects: the datasets could impact the models along dimensions that may seem unrelated to the intended tasks. Our state-vector framework paves the way for a systematic understanding of the dataset effects, a crucial component in responsible and robust model development.</abstract>
    <identifier type="citekey">sahak-etal-2023-state</identifier>
    <identifier type="doi">10.18653/v1/2023.emnlp-main.942</identifier>
    <location>
        <url>https://aclanthology.org/2023.emnlp-main.942</url>
    </location>
    <part>
        <date>2023-12</date>
        <extent unit="page">
            <start>15231</start>
            <end>15245</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T A State-Vector Framework for Dataset Effects
%A Sahak, Esmat
%A Zhu, Zining
%A Rudzicz, Frank
%Y Bouamor, Houda
%Y Pino, Juan
%Y Bali, Kalika
%S Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing
%D 2023
%8 December
%I Association for Computational Linguistics
%C Singapore
%F sahak-etal-2023-state
%X The impressive success of recent deep neural network (DNN)-based systems is significantly influenced by the high-quality datasets used in training. However, the effects of the datasets, especially how they interact with each other, remain underexplored. We propose a state-vector framework to enable rigorous studies in this direction. This framework uses idealized probing test results as the bases of a vector space. This framework allows us to quantify the effects of both standalone and interacting datasets. We show that the significant effects of some commonly-used language understanding datasets are characteristic and are concentrated on a few linguistic dimensions. Additionally, we observe some “spill-over” effects: the datasets could impact the models along dimensions that may seem unrelated to the intended tasks. Our state-vector framework paves the way for a systematic understanding of the dataset effects, a crucial component in responsible and robust model development.
%R 10.18653/v1/2023.emnlp-main.942
%U https://aclanthology.org/2023.emnlp-main.942
%U https://doi.org/10.18653/v1/2023.emnlp-main.942
%P 15231-15245

Markdown (Informal)

[A State-Vector Framework for Dataset Effects](https://aclanthology.org/2023.emnlp-main.942) (Sahak et al., EMNLP 2023)

• A State-Vector Framework for Dataset Effects (Sahak et al., EMNLP 2023)

ACL

• Esmat Sahak, Zining Zhu, and Frank Rudzicz. 2023. A State-Vector Framework for Dataset Effects. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 15231–15245, Singapore. Association for Computational Linguistics.