Egy emBERT próbáló feladat
Dávid Márk Nemeskey
In XVI. Magyar Számítógépes Nyelvészeti Konferencia,
2020
PDF
In the last couple of years deep, contextual embeddings have superseded traditional, manually compiled feature sets in most NLP tasks. However, the Hungarian NLP pipelines (e-magyar, magyarlanc) still manual features. In this article, we introduce the emBERT module, which allows the integration of contextual embedding-based classifiers into e-magyar, via the transformers library.
The module provides classifiers for named netity recognition and NP chunking, achieving state-of-the-art performance.
Citation
@InProceedings{ Nemeskey:2020a,
author = {Nemeskey, Dávid Márk},
title = {Egy \texttt{emBERT} próbáló feladat},
booktitle = {{XVI}.\ Magyar Sz{\'a}m{\'i}t{\'o}g{\'e}pes Nyelv{\'e}szeti Konferencia ({MSZNY}2020)},
year = 2020,
pages = {409--418},
address = {Szeged},
}