Mimicking Neural Machine Translation History for Pedagogic Reasons

Vincent Vandeghinste

Mimicking Neural Machine Translation History for Pedagogic Reasons

Correct Metadata for

Use this form to create a GitHub issue with structured data describing the correction. You will need a GitHub account. Once you create that issue, the correction will be reviewed by a staff member.

⚠️ Mobile Users: Submitting this form to create a new issue will only work with github.com, not the GitHub Mobile app.

Important: The Anthology treat PDFs as authoritative. Please use this form only to correct data that is out of line with the PDF. See our corrections guidelines if you need to change the PDF.

Title Adjust the title. Retain tags such as <fixed-case>.

Authors Adjust author names and order to match the PDF.

Abstract Correct abstract if needed. Retain XML formatting tags such as <tex-math>. You may use ... for bold, ... for italic, ... for underline, <sc>...</sc> for small-caps, <tt>...<tt> for typewriter text, <url>...</url> for URLs, <a href=...> for hyperlinks, and <par/> for paragraph breaks.

Verification against PDF Ensure that the new title/authors match the snapshot below. (If there is no snapshot or it is too small, consult the PDF.)

Authors concatenated from the text boxes above:

ALL author names match the snapshot above—including middle initials, hyphens, and accents.

Abstract

In this paper we describe how we mimick the different steps in the historical development of NMT systems, all trained and evaluated on the same small data set. We do this for pedagogic reasons so students can see the effect of each of the steps on metrics like BLEU but also on qualitative examples, which the training scripts generate after each epoch. As MT paradigms, we discuss NMT training from scratch, finetuning pretrained encoder-decoder models, and finally prompt engineering for decoder only models. All models run in Kaggle sessions and all Python scripts and JuPyter notebooks are made available to the MT teaching community through Github and public Kaggle sessions.

Anthology ID:: 2026.taitt-1.3
Volume:: Proceedings of the 1st International Workshop on Teaching AI-Based Translation and Technologies (TAITT 2026)
Month:: June
Year:: 2026
Address:: Tilburg, the Netherlands
Editors:: Ralph Krüger, Dorothy Kenny, Sheila Castilho, Sergi Álvarez-Vidal, Nora Aranberri, María Isabel Rivas Ginel, Janiça Hackenbuchner
Venues:: TAITT | WS
SIG:
Publisher:: European Association for Machine Translation
Note:
Pages:: 19–27
Language:
URL:: https://aclanthology.org/2026.taitt-1.3/
DOI:
Bibkey:
Cite (ACL):: Vincent Vandeghinste. 2026. Mimicking Neural Machine Translation History for Pedagogic Reasons. In Proceedings of the 1st International Workshop on Teaching AI-Based Translation and Technologies (TAITT 2026), pages 19–27, Tilburg, the Netherlands. European Association for Machine Translation.
Cite (Informal):: Mimicking Neural Machine Translation History for Pedagogic Reasons (Vandeghinste, TAITT 2026)
Copy Citation:
PDF:: https://aclanthology.org/2026.taitt-1.3.pdf

PDF Cite Search Fix data

Export citation

BibTeX
MODS XML
Endnote
Preformatted

@inproceedings{vandeghinste-2026-mimicking,
    title = "Mimicking Neural Machine Translation History for Pedagogic Reasons",
    author = "Vandeghinste, Vincent",
    editor = {Kr{\"u}ger, Ralph  and
      Kenny, Dorothy  and
      Castilho, Sheila  and
      {\'A}lvarez-Vidal, Sergi  and
      Aranberri, Nora  and
      Ginel, Mar{\'i}a Isabel Rivas  and
      Hackenbuchner, Jani{\c{c}}a},
    booktitle = "Proceedings of the 1st International Workshop on Teaching {AI}-Based Translation and Technologies ({TAITT} 2026)",
    month = jun,
    year = "2026",
    address = "Tilburg, the Netherlands",
    publisher = "European Association for Machine Translation",
    url = "https://aclanthology.org/2026.taitt-1.3/",
    pages = "19--27",
    abstract = "In this paper we describe how we mimick the different steps in the historical development of NMT systems, all trained and evaluated on the same small data set. We do this for pedagogic reasons so students can see the effect of each of the steps on metrics like BLEU but also on qualitative examples, which the training scripts generate after each epoch. As MT paradigms, we discuss NMT training from scratch, finetuning pretrained encoder-decoder models, and finally prompt engineering for decoder only models. All models run in Kaggle sessions and all Python scripts and JuPyter notebooks are made available to the MT teaching community through Github and public Kaggle sessions."
}

Download as File

<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="vandeghinste-2026-mimicking">
    <titleInfo>
        <title>Mimicking Neural Machine Translation History for Pedagogic Reasons</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Vincent</namePart>
        <namePart type="family">Vandeghinste</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-06</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 1st International Workshop on Teaching AI-Based Translation and Technologies (TAITT 2026)</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Ralph</namePart>
            <namePart type="family">Krüger</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Dorothy</namePart>
            <namePart type="family">Kenny</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Sheila</namePart>
            <namePart type="family">Castilho</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Sergi</namePart>
            <namePart type="family">Álvarez-Vidal</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Nora</namePart>
            <namePart type="family">Aranberri</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">María</namePart>
            <namePart type="given">Isabel</namePart>
            <namePart type="given">Rivas</namePart>
            <namePart type="family">Ginel</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Janiça</namePart>
            <namePart type="family">Hackenbuchner</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>European Association for Machine Translation</publisher>
            <place>
                <placeTerm type="text">Tilburg, the Netherlands</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>In this paper we describe how we mimick the different steps in the historical development of NMT systems, all trained and evaluated on the same small data set. We do this for pedagogic reasons so students can see the effect of each of the steps on metrics like BLEU but also on qualitative examples, which the training scripts generate after each epoch. As MT paradigms, we discuss NMT training from scratch, finetuning pretrained encoder-decoder models, and finally prompt engineering for decoder only models. All models run in Kaggle sessions and all Python scripts and JuPyter notebooks are made available to the MT teaching community through Github and public Kaggle sessions.</abstract>
    <identifier type="citekey">vandeghinste-2026-mimicking</identifier>
    <location>
        <url>https://aclanthology.org/2026.taitt-1.3/</url>
    </location>
    <part>
        <date>2026-06</date>
        <extent unit="page">
            <start>19</start>
            <end>27</end>
        </extent>
    </part>
</mods>
</modsCollection>

Download as File

%0 Conference Proceedings
%T Mimicking Neural Machine Translation History for Pedagogic Reasons
%A Vandeghinste, Vincent
%Y Krüger, Ralph
%Y Kenny, Dorothy
%Y Castilho, Sheila
%Y Álvarez-Vidal, Sergi
%Y Aranberri, Nora
%Y Ginel, María Isabel Rivas
%Y Hackenbuchner, Janiça
%S Proceedings of the 1st International Workshop on Teaching AI-Based Translation and Technologies (TAITT 2026)
%D 2026
%8 June
%I European Association for Machine Translation
%C Tilburg, the Netherlands
%F vandeghinste-2026-mimicking
%X In this paper we describe how we mimick the different steps in the historical development of NMT systems, all trained and evaluated on the same small data set. We do this for pedagogic reasons so students can see the effect of each of the steps on metrics like BLEU but also on qualitative examples, which the training scripts generate after each epoch. As MT paradigms, we discuss NMT training from scratch, finetuning pretrained encoder-decoder models, and finally prompt engineering for decoder only models. All models run in Kaggle sessions and all Python scripts and JuPyter notebooks are made available to the MT teaching community through Github and public Kaggle sessions.
%U https://aclanthology.org/2026.taitt-1.3/
%P 19-27

Download as File

Markdown (Informal)

[Mimicking Neural Machine Translation History for Pedagogic Reasons](https://aclanthology.org/2026.taitt-1.3/) (Vandeghinste, TAITT 2026)

Mimicking Neural Machine Translation History for Pedagogic Reasons (Vandeghinste, TAITT 2026)

ACL

Vincent Vandeghinste. 2026. Mimicking Neural Machine Translation History for Pedagogic Reasons. In Proceedings of the 1st International Workshop on Teaching AI-Based Translation and Technologies (TAITT 2026), pages 19–27, Tilburg, the Netherlands. European Association for Machine Translation.