@inproceedings{dankers-etal-2023-memorisation,
title = "Memorisation Cartography: Mapping out the Memorisation-Generalisation Continuum in Neural Machine Translation",
author = "Dankers, Verna and
Titov, Ivan and
Hupkes, Dieuwke",
editor = "Bouamor, Houda and
Pino, Juan and
Bali, Kalika",
booktitle = "Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing",
month = dec,
year = "2023",
address = "Singapore",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.emnlp-main.518",
doi = "10.18653/v1/2023.emnlp-main.518",
pages = "8323--8343",
abstract = "When training a neural network, it will quickly memorise some source-target mappings from your dataset but never learn some others. Yet, memorisation is not easily expressed as a binary feature that is good or bad: individual datapoints lie on a memorisation-generalisation continuum. What determines a datapoint{'}s position on that spectrum, and how does that spectrum influence neural models{'} performance? We address these two questions for neural machine translation (NMT) models. We use the counterfactual memorisation metric to (1) build a resource that places 5M NMT datapoints on a memorisation-generalisation map, (2) illustrate how the datapoints{'} surface-level characteristics and a models{'} per-datum training signals are predictive of memorisation in NMT, (3) and describe the influence that subsets of that map have on NMT systems{'} performance.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="dankers-etal-2023-memorisation">
<titleInfo>
<title>Memorisation Cartography: Mapping out the Memorisation-Generalisation Continuum in Neural Machine Translation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Verna</namePart>
<namePart type="family">Dankers</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ivan</namePart>
<namePart type="family">Titov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dieuwke</namePart>
<namePart type="family">Hupkes</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>When training a neural network, it will quickly memorise some source-target mappings from your dataset but never learn some others. Yet, memorisation is not easily expressed as a binary feature that is good or bad: individual datapoints lie on a memorisation-generalisation continuum. What determines a datapoint’s position on that spectrum, and how does that spectrum influence neural models’ performance? We address these two questions for neural machine translation (NMT) models. We use the counterfactual memorisation metric to (1) build a resource that places 5M NMT datapoints on a memorisation-generalisation map, (2) illustrate how the datapoints’ surface-level characteristics and a models’ per-datum training signals are predictive of memorisation in NMT, (3) and describe the influence that subsets of that map have on NMT systems’ performance.</abstract>
<identifier type="citekey">dankers-etal-2023-memorisation</identifier>
<identifier type="doi">10.18653/v1/2023.emnlp-main.518</identifier>
<location>
<url>https://aclanthology.org/2023.emnlp-main.518</url>
</location>
<part>
<date>2023-12</date>
<extent unit="page">
<start>8323</start>
<end>8343</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Memorisation Cartography: Mapping out the Memorisation-Generalisation Continuum in Neural Machine Translation
%A Dankers, Verna
%A Titov, Ivan
%A Hupkes, Dieuwke
%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 dankers-etal-2023-memorisation
%X When training a neural network, it will quickly memorise some source-target mappings from your dataset but never learn some others. Yet, memorisation is not easily expressed as a binary feature that is good or bad: individual datapoints lie on a memorisation-generalisation continuum. What determines a datapoint’s position on that spectrum, and how does that spectrum influence neural models’ performance? We address these two questions for neural machine translation (NMT) models. We use the counterfactual memorisation metric to (1) build a resource that places 5M NMT datapoints on a memorisation-generalisation map, (2) illustrate how the datapoints’ surface-level characteristics and a models’ per-datum training signals are predictive of memorisation in NMT, (3) and describe the influence that subsets of that map have on NMT systems’ performance.
%R 10.18653/v1/2023.emnlp-main.518
%U https://aclanthology.org/2023.emnlp-main.518
%U https://doi.org/10.18653/v1/2023.emnlp-main.518
%P 8323-8343
Markdown (Informal)
[Memorisation Cartography: Mapping out the Memorisation-Generalisation Continuum in Neural Machine Translation](https://aclanthology.org/2023.emnlp-main.518) (Dankers et al., EMNLP 2023)
ACL