1 |
From FreEM to D'AlemBERT ; From FreEM to D'AlemBERT: a Large Corpus and a Language Model for Early Modern French
|
|
|
|
In: Proceedings of the 13th Language Resources and Evaluation Conference ; https://hal.inria.fr/hal-03596653 ; Proceedings of the 13th Language Resources and Evaluation Conference, European Language Resources Association, Jun 2022, Marseille, France (2022)
|
|
Abstract:
8 pages, 2 figures, 4 tables ; International audience ; Language models for historical states of language are becoming increasingly important to allow the optimal digitisation and analysis of old textual sources. Because these historical states are at the same time more complex to process and more scarce in the corpora available, specific efforts are necessary to train natural language processing (NLP) tools adapted to the data. In this paper, we present our efforts to develop NLP tools for Early Modern French (historical French from the 16th to the 18th centuries). We present the FreEMmax corpus of Early Modern French and D'AlemBERT, a RoBERTa-based language model trained on FreEMmax. We evaluate the usefulness of D'AlemBERT by fine-tuning it on a part-of-speech tagging task, outperforming previous work on the test set. Importantly, we find evidence for the transfer learning capacity of the language model, since its performance on lesser-resourced time periods appears to have been boosted by the more resourced ones. We release D'AlemBERT and the open-sourced subpart of the FreEMmax corpus.
|
|
Keyword:
[INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL]; Corpus creation; Création de corpus; Digital humanities; Early Modern French; Français classique; Humanités Numériques; Language modelling; Langues peu dotées; Less-resourced languages; Modèle de langue neuronal; Modélisation linguistique; Neural language representation models; Partie du discours; POS tagging
|
|
URL: https://hal.inria.fr/hal-03596653
|
|
BASE
|
|
Hide details
|
|
2 |
A dataset for automatic detection of places in (early) modern French texts ; Un jeu de données pour la détection automatique de lieux dans les textes français modernes
|
|
|
|
In: NASSCFL 2021 - 50th Annual North American Society for Seventeenth-Century French Literature Conference ; https://hal.archives-ouvertes.fr/hal-03187097 ; NASSCFL 2021 - 50th Annual North American Society for Seventeenth-Century French Literature Conference, NASSCFL, May 2021, Iowa City / Virtual, United States. pp.5 (2021)
|
|
BASE
|
|
Show details
|
|
3 |
Expanding the content model of annotationBlock
|
|
|
|
In: Next Gen TEI, 2021 - TEI Conference and Members’ Meeting ; https://hal.archives-ouvertes.fr/hal-03380805 ; Next Gen TEI, 2021 - TEI Conference and Members’ Meeting, Oct 2021, Virtual, United States (2021)
|
|
BASE
|
|
Show details
|
|
|
|