Doubly-Trained Adversarial Data Augmentation for Neural Machine Translation

10/12/2021
by   Weiting Tan, et al.
0

Neural Machine Translation (NMT) models are known to suffer from noisy inputs. To make models robust, we generate adversarial augmentation samples that attack the model and preserve the source-side semantic meaning at the same time. To generate such samples, we propose a doubly-trained architecture that pairs two NMT models of opposite translation directions with a joint loss function, which combines the target-side attack and the source-side semantic similarity constraint. The results from our experiments across three different language pairs and two evaluation metrics show that these adversarial samples improve the model robustness.

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset