GETALP's repositories
disambiguate
Disambiguate is a tool for training and using state of the art neural WSD models
mass-dataset
MaSS - Multilingual corpus of Sentence-aligned Spoken utterances
Lightweight-Transformer-Models-For-HAR-on-Mobile-Devices
Human Activity Recognition Transformer (HART) is a transformer based architecture that has been specifically adapted for IMU sensing devices. Findings shows that HART uses fewer paremeters, FLOPs and achieves state-of-the-art results.
PerCom2021-FL
Federated_learning codes used for the the paper "Evaluation of Federated Learning Aggregation Algorithms" and "A Federated Learning Aggregation Algorithm for Pervasive Computing: Evaluation and Comparison"
genderednews
Code of the GenderedNews project
Arasaac-WN
Links from Arasaac pictogram database to the Princeton WordNet
WCE-SLT-LIG
Corpus for evaluation of WCE in Spoken Language Translation
Cross-Language-Dataset
A multilingual, multi-style and multi-granularity dataset for cross-language textual similarity detection
disambiguate-translate
[Work In Progress] Disambiguate-Translate is an extension of the "Disambiguate" tool for jointly learning to disambiguate and translate text at the same time.
WSD-TALN2016-Hadjsalahetal
French corpus annotated with WordNet 3.0 tags. (see article Amélioration de la traduction automatique d’un corpus annoté - TALN 2017 (Marwa Hadj Salah, Hervé Blanchon, Mounir Zrigui, Didier Schwab)