Módszerek összehasonlítása

	Multilingvális GAN	Multilingvis mondatbeágyazások
Tudományterület	Mélytanulás	Mélytanulás
Módszercsalád	Machine learning	Machine learning
Keletkezés éve	2017–2019	2019–2022
Megalkotó	Goodfellow et al. (GAN); multilingual extensions by various authors from 2017 onward	Reimers, N. & Gurevych, I.; Feng, F. et al. (Google)
Típus	Generative adversarial model with multilingual conditioning	Cross-lingual representation learning
Alapmű	Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., & Bengio, Y. (2014). Generative Adversarial Nets. Advances in Neural Information Processing Systems (NeurIPS), 27.	Reimers, N. & Gurevych, I. (2020). Making Monolingual Sentence Embeddings Multilingual using Knowledge Distillation. Proceedings of EMNLP 2020, 4512–4525.
Alternatív nevek	Multilingual GAN, Cross-lingual GAN, Multilingual Generative Adversarial Network, ML-GAN	multilingual sentence representations, cross-lingual sentence embeddings, mSE, multilingual semantic embeddings
Kapcsolódó	5	5
Összefoglaló	A Multilingual GAN pairs the generative adversarial framework with cross-lingual components — a shared encoder, language-conditioned generator, and a language discriminator — so that a single model can generate or align representations across multiple languages simultaneously. It is applied to cross-lingual text generation, machine translation, multilingual data augmentation, and language-invariant feature learning.	Multilingual sentence embeddings map sentences from many languages into a single shared vector space so that semantically equivalent sentences — regardless of language — land close together. Models such as LaBSE, multilingual Sentence-BERT, and mUSE have made it practical to compare, retrieve, and classify text across 50 to 100+ languages without translating anything first.