DOLFIN -- Document-Level Financial test set for Machine Translation

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Nakhlé, Mariam, Dinarelli, Marco, Qader, Raheel, Esperança-Rodier, Emmanuelle, Blanchon, Hervé
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866909477716885504
author Nakhlé, Mariam
Dinarelli, Marco
Qader, Raheel
Esperança-Rodier, Emmanuelle
Blanchon, Hervé
author_facet Nakhlé, Mariam
Dinarelli, Marco
Qader, Raheel
Esperança-Rodier, Emmanuelle
Blanchon, Hervé
contents Despite the strong research interest in document-level Machine Translation (MT), the test sets dedicated to this task are still scarce. The existing test sets mainly cover topics from the general domain and fall short on specialised domains, such as legal and financial. Also, in spite of their document-level aspect, they still follow a sentence-level logic that does not allow for including certain linguistic phenomena such as information reorganisation. In this work, we aim to fill this gap by proposing a novel test set: DOLFIN. The dataset is built from specialised financial documents, and it makes a step towards true document-level MT by abandoning the paradigm of perfectly aligned sentences, presenting data in units of sections rather than sentences. The test set consists of an average of 1950 aligned sections for five language pairs. We present a detailed data collection pipeline that can serve as inspiration for aligning new document-level datasets. We demonstrate the usefulness and quality of this test set by evaluating a number of models. Our results show that the test set is able to discriminate between context-sensitive and context-agnostic models and shows the weaknesses when models fail to accurately translate financial texts. The test set is made public for the community.
format Preprint
id arxiv_https___arxiv_org_abs_2502_03053
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DOLFIN -- Document-Level Financial test set for Machine Translation
Nakhlé, Mariam
Dinarelli, Marco
Qader, Raheel
Esperança-Rodier, Emmanuelle
Blanchon, Hervé
Computation and Language
Despite the strong research interest in document-level Machine Translation (MT), the test sets dedicated to this task are still scarce. The existing test sets mainly cover topics from the general domain and fall short on specialised domains, such as legal and financial. Also, in spite of their document-level aspect, they still follow a sentence-level logic that does not allow for including certain linguistic phenomena such as information reorganisation. In this work, we aim to fill this gap by proposing a novel test set: DOLFIN. The dataset is built from specialised financial documents, and it makes a step towards true document-level MT by abandoning the paradigm of perfectly aligned sentences, presenting data in units of sections rather than sentences. The test set consists of an average of 1950 aligned sections for five language pairs. We present a detailed data collection pipeline that can serve as inspiration for aligning new document-level datasets. We demonstrate the usefulness and quality of this test set by evaluating a number of models. Our results show that the test set is able to discriminate between context-sensitive and context-agnostic models and shows the weaknesses when models fail to accurately translate financial texts. The test set is made public for the community.
title DOLFIN -- Document-Level Financial test set for Machine Translation
topic Computation and Language
url https://arxiv.org/abs/2502.03053