Многоязычная Тонкая Настройка
Related Papers
Can Monolingual Pretrained Models Help Cross-Lingual Classification?
AbstractMultilingual pretrained language models (such as multilingual BERT) have achieved impressive results for cross-lingual transfer. However, due to the constant model capacity, multilingual pre-training usually lags behind the monolingual competitors. …...
Don’t Go Far Off: An Empirical Study on Neural Poetry Translation
AbstractDespite constant improvements in machine translation quality, automatic poetry translation remains a challenging problem due to the lack of open-sourced parallel poetic corpora, and to the intrinsic complexities involved in …...
Multilingual Translation from Denoising Pre-Training
Recent work demonstrates the potential of training one model for multilingual machine translation. In parallel, denoising pretraining using unlabeled monolingual data as a starting point for finetuning bitext machine translation …...
Towards Making the Most of Cross-Lingual Transfer for Zero-Shot Neural Machine Translation
AbstractThis paper demonstrates that multilingual pretraining and multilingual fine-tuning are both critical for facilitating cross-lingual transfer in zero-shot translation, where the neural machine translation (NMT) model is tested on source …...
Improving Low-Resource Speech Recognition through Multilingual Fine-Tuning with Language Identifiers and Self-Training
Previous work has demonstrated that multilin- gual fine-tuning of a pretrained multilingual speech representation model can lead to im- proved speech recognition accuracy when there is extremely little target language …...