Secoco: Self-Correcting Encoding for Neural Machine Translation
This addresses robustness issues in NMT for noisy inputs, offering a novel method that is incremental but with strong specific gains.
The paper tackles input noise in neural machine translation by introducing Secoco, a self-correcting encoding framework that corrects and deletes errors during decoding, achieving significant improvements on real-world and benchmark datasets.
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 over strong baselines on two real-world test sets and a benchmark WMT dataset with good interpretability. We will make our code and dataset publicly available soon.