Most existing methods on robust neural machine translation (NMT) construct adversarial examples by injecting noise into authentic examples and indiscriminately exploit two types of examples.
In this paper, we propose Regularized Contrastive Cross-lingual Cross-modal (RC^3) pre-training, which further exploits more abundant weakly-aligned multilingual image-text pairs.
To address these issues, in this paper, we propose a multi-task multi-stage transitional (MMT) training framework, where an NCT model is trained using the bilingual chat translation dataset and additional monolingual dialogues.
Simile recognition involves two subtasks: simile sentence classification that discriminates whether a sentence contains simile, and simile component extraction that locates the corresponding objects (i. e., tenors and vehicles).
Meanwhile we inject two types of perturbations into the retrieved pairs for robust training.
The goal of the cross-lingual summarization (CLS) is to convert a document in one language (e. g., English) to a summary in another one (e. g., Chinese).
Most dominant neural machine translation (NMT) models are restricted to make predictions only according to the local context of preceding words in a left-to-right manner.
Neural Chat Translation (NCT) aims to translate conversational text between speakers of different languages.
Due to the great potential in facilitating software development, code generation has attracted increasing attention recently.
Multi-modal neural machine translation (NMT) aims to translate source sentences into a target language paired with images.
Finally, we combine the discourse structure information with the word embedding before it is fed into the encoder.
Unsupervised style transfer aims to change the style of an input sentence while preserving its original content without using parallel training data.