<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="bolliger-etal-2023-scandl">
    <titleInfo>
        <title>ScanDL: A Diffusion Model for Generating Synthetic Scanpaths on Texts</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Lena</namePart>
        <namePart type="family">Bolliger</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">David</namePart>
        <namePart type="family">Reich</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Patrick</namePart>
        <namePart type="family">Haller</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Deborah</namePart>
        <namePart type="family">Jakobi</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Paul</namePart>
        <namePart type="family">Prasse</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Lena</namePart>
        <namePart type="family">Jäger</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>Eye movements in reading play a crucial role in psycholinguistic research studying the cognitive mechanisms underlying human language processing. More recently, the tight coupling between eye movements and cognition has also been leveraged for language-related machine learning tasks such as the interpretability, enhancement, and pre-training of language models, as well as the inference of reader- and text-specific properties. However, scarcity of eye movement data and its unavailability at application time poses a major challenge for this line of research. Initially, this problem was tackled by resorting to cognitive models for synthesizing eye movement data. However, for the sole purpose of generating human-like scanpaths, purely data-driven machine-learning-based methods have proven to be more suitable. Following recent advances in adapting diffusion processes to discrete data, we propose ScanDL, a novel discrete sequence-to-sequence diffusion model that generates synthetic scanpaths on texts. By leveraging pre-trained word representations and jointly embedding both the stimulus text and the fixation sequence, our model captures multi-modal interactions between the two inputs. We evaluate ScanDL within- and across-dataset and demonstrate that it significantly outperforms state-of-the-art scanpath generation methods. Finally, we provide an extensive psycholinguistic analysis that underlines the model’s ability to exhibit human-like reading behavior. Our implementation is made available at https://github.com/DiLi-Lab/ScanDL.</abstract>
    <identifier type="citekey">bolliger-etal-2023-scandl</identifier>
    <identifier type="doi">10.18653/v1/2023.emnlp-main.960</identifier>
    <location>
        <url>https://aclanthology.org/2023.emnlp-main.960</url>
    </location>
    <part>
        <date>2023-12</date>
        <extent unit="page">
            <start>15513</start>
            <end>15538</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T ScanDL: A Diffusion Model for Generating Synthetic Scanpaths on Texts
%A Bolliger, Lena
%A Reich, David
%A Haller, Patrick
%A Jakobi, Deborah
%A Prasse, Paul
%A Jäger, Lena
%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 bolliger-etal-2023-scandl
%X Eye movements in reading play a crucial role in psycholinguistic research studying the cognitive mechanisms underlying human language processing. More recently, the tight coupling between eye movements and cognition has also been leveraged for language-related machine learning tasks such as the interpretability, enhancement, and pre-training of language models, as well as the inference of reader- and text-specific properties. However, scarcity of eye movement data and its unavailability at application time poses a major challenge for this line of research. Initially, this problem was tackled by resorting to cognitive models for synthesizing eye movement data. However, for the sole purpose of generating human-like scanpaths, purely data-driven machine-learning-based methods have proven to be more suitable. Following recent advances in adapting diffusion processes to discrete data, we propose ScanDL, a novel discrete sequence-to-sequence diffusion model that generates synthetic scanpaths on texts. By leveraging pre-trained word representations and jointly embedding both the stimulus text and the fixation sequence, our model captures multi-modal interactions between the two inputs. We evaluate ScanDL within- and across-dataset and demonstrate that it significantly outperforms state-of-the-art scanpath generation methods. Finally, we provide an extensive psycholinguistic analysis that underlines the model’s ability to exhibit human-like reading behavior. Our implementation is made available at https://github.com/DiLi-Lab/ScanDL.
%R 10.18653/v1/2023.emnlp-main.960
%U https://aclanthology.org/2023.emnlp-main.960
%U https://doi.org/10.18653/v1/2023.emnlp-main.960
%P 15513-15538

Markdown (Informal)

[ScanDL: A Diffusion Model for Generating Synthetic Scanpaths on Texts](https://aclanthology.org/2023.emnlp-main.960) (Bolliger et al., EMNLP 2023)

• ScanDL: A Diffusion Model for Generating Synthetic Scanpaths on Texts (Bolliger et al., EMNLP 2023)

ACL

• Lena Bolliger, David Reich, Patrick Haller, Deborah Jakobi, Paul Prasse, and Lena Jäger. 2023. ScanDL: A Diffusion Model for Generating Synthetic Scanpaths on Texts. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 15513–15538, Singapore. Association for Computational Linguistics.