Comparer des méthodes

	Réseau de neurones convolutif multilingue	Transformeur multilingue
Domaine	Apprentissage profond	Apprentissage profond
Famille	Machine learning	Machine learning
Année d'origine	2014–2016	2019–2020
Auteur d'origine	Kim, Y. (seminal NLP CNN); multilingual extension by community	Devlin et al. (mBERT); Conneau et al. (XLM-R)
Type	Deep learning classifier	Pre-trained cross-lingual language model
Source fondatrice	Kim, Y. (2014). Convolutional Neural Networks for Sentence Classification. Proceedings of EMNLP 2014, pp. 1746–1751.	Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019). BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. Proceedings of NAACL-HLT 2019, pp. 4171–4186. Association for Computational Linguistics.
Alias	ML-CNN, cross-lingual CNN, multilingual text CNN, multilingual ConvNet	multilingual LM, cross-lingual transformer, mBERT-style model, multilingual pre-trained model
Apparentées	4	4
Résumé	A Multilingual CNN applies convolutional filters over token embeddings drawn from two or more languages, producing shared feature representations that enable a single model to classify, tag, or extract information across language boundaries without training separate models per language. It extends the standard text-CNN architecture with multilingual or cross-lingual input embeddings.	A multilingual transformer is a pre-trained language model built on the transformer architecture and trained jointly on text from dozens to over one hundred languages. Models such as mBERT and XLM-RoBERTa learn shared cross-lingual representations, enabling zero-shot or few-shot transfer: a model fine-tuned on English data can often be applied directly to French, German, Arabic, or Chinese without language-specific labels.