Secoco: Self-Correcting Encoding for Neural Machine Translation

Abstract

This paper presents Self-correcting Encoding (Secoco), a framework that effectively deals with input noise for robust neural machine translation by introducing self-correcting predictors. Different from previous robust approaches, Secoco enables NMT to explicitly correct noisy inputs and delete specific errors simultaneously with the translation decoding process. Secoco is able to achieve significant improvements of 1.6 BLEU points over strong baselines on two real-world test sets and a benchmark WMT dataset with good interpretability.The code and dataset are publicly available at https://github.com/rgwt123/Secoco.

Experiences of Adapting Multimodal Machine Translation Techniques for Hindi

“…Other methods include the construction of source and target side adversarial examples [11] for noisy NMT. Subsequently, Insertion and Deletion predictors were added to the NMT, which improved results on noisy inputs [73]. It was observed that multi-tasking with translation and denoising objectives helps to generate robust translation [56].…”