Accéder directement au contenu

Arthur Amalvy

PhD student at Laboratoire Informatique d'Avignon (Avignon Computer Science Lab)
6
Documents
Affiliations actuelles
Identifiants chercheurs
  • IdHAL aamalvy

Présentation

I am currently a PhD Student at [Laboratoire Informatique d'Avignon](https://lia.univ-avignon.fr/) under the supervision of Vincent Labatut and Richard Dufour. My current research interests are centered on character networks extraction in novels. This includes a number of natural langage processing subtasks: named entity recognition, coreference resolution, alias resolution, quote detection, speaker attribution... #### Online Identities - [Google Scholar](https://scholar.google.com/citations?user=yHOkn78AAAAJ&hl=fr&oi=ao) - [ORCID](https://orcid.org/0000-0003-4629-0923) ### Education - **2021-now, Laboratoire Informatique d'Avignon:** PhD in Natural Language Processing - **2019-2020, National Central University, Taiwan :** Master of Computer Science and Information Engineering - **2017-2020, Université de Technologie de Belfort-Montbéliard, France :** Engineer's Degree in Computer Science - **2015-2020, Université de Technologie de Belfort-Montbéliard, France :** DEUTEC ### Teaching #### 2021-2022 - Basics of Programming level 1 (C++), L1 - Internet Design (HTML/CSS), L1 #### 2022-2023 - Basics of Programming level 2 (C++), L1 - Prototyping and User Interfaces (JavaFX), M1 - Mobile Applications (Android, Kotlin), L3 #### 2023-2024 - Application of Innovation (Research Oriented), M2 - Basics of Programming level 2 (C++), L1 - Prototyping and User Interfaces (JavaFX), M1 ### Previous Industry Experience - **2020-2021, Prokov Editions, Nancy :** Artificial Intelligence R&amp;D Engineer - **2018, Idemia, Cergy :** Cloud Microservices Architect Intern ### Softwares - **[Renard](https://github.com/CompNet/Renard):** Renard (Relationships Extraction from NARrative Documents) is a modular Python library to extract static and dynamic character networks, made as part of my PhD. - **[Tibert](https://github.com/CompNet/Tibert):** A simple to use coreference resolution library, created to support Renard. - **[blink](https://gitlab.com/sharpattack/blink):** A Python library to work with RPLidar A2M8. - ... and a lot of other stuff! See my [Github](https://github.com/Aethor) and my [Gitlab](https://gitlab.com/Aethor). ### Other I also participate to the French robotics cup every year with my team https://sharpattack.fr/en/ #### Contact Informations - Email : <arthur.amalvy@univ-avignon.fr>

Publications

richard-dufour
Image document

Learning to Rank Context for Named Entity Recognition Using a Synthetic Dataset

Arthur Amalvy , Vincent Labatut , Richard Dufour
Conference on Empirical Methods in Natural Language Processing (EMNLP), ACL, Dec 2023, Singapore, Singapore. pp.10372-10382
Communication dans un congrès hal-04237987v2
Image document

The Role of Global and Local Context in Named Entity Recognition

Arthur Amalvy , Vincent Labatut , Richard Dufour
61st Annual Meeting of the Association for Computational Linguistics (ACL), Association for Computational Linguistics, Jul 2023, Toronto, Canada. pp.714-722, ⟨10.18653/v1/2023.acl-short.62⟩
Communication dans un congrès hal-04092431v2
Image document

BERT meets d'Artagnan: Data Augmentation for Robust Character Detection in Novels

Arthur Amalvy , Vincent Labatut , Richard Dufour
Workshop on Computational Methods in the Humanities (COMHUM), Jun 2022, Lausanne, Switzerland
Communication dans un congrès hal-03617722v2
Image document

Data Augmentation for Robust Character Detection in Fantasy Novels

Arthur Amalvy , Vincent Labatut , Richard Dufour
Workshop on Computational Methods in the Humanities 2022, Jun 2022, Lausanne, Switzerland
Communication dans un congrès hal-03972448v1
Image document

Remplacement de mentions pour l'adaptation d'un corpus de reconnaissance d'entités nommées à un domaine cible

Arthur Amalvy , Vincent Labatut , Richard Dufour
29ème Conférence sur le Traitement Automatique des Langues Naturelles (TALN), Jun 2022, Avignon, France. pp.198-205
Communication dans un congrès hal-03651510v3