Our paper was accepted for NAACL 2022 (main).
Machel Reid and Mikel Artetxe “PARADISE: Exploiting Parallel Data for Multilingual Sequence-to-Sequence Pretraining”. The 2022 Conference of the North American Chapter of the Association for Computational Linguistics (NAACL 2022). July 2022. Association for Computational Linguistics.
Despite the success of multilingual sequence-to-sequence pretraining, most existing approaches rely on monolingual corpora, and do not make use of the strong cross-lingual signal contained in parallel data. In this paper, we present PARADISE (PARAllel & Denoising Integration in SEquence-to-sequence models), which extends the conventional denoising objective used to train these models by (i) replacing words in the noised sequence according to a multilingual dictionary, and (ii) predicting the reference translation according to a parallel corpus instead of recovering the original sequence. Our experiments on machine translation and cross-lingual natural language inference show an average improvement of 2.0 BLEU points and 6.7 accuracy points from integrating parallel data into pretraining, respectively, obtaining results that are competitive with several popular models at a fraction of their computational cost.