Papers › Counterfactual Data Augmentation for Neural Machine Translation

Counterfactual Data Augmentation for Neural Machine Translation

1 Jun 2021NAACL 2021 4archive 2025-07-28

Qi Liu, Matt Kusner, Phil Blunsom

We propose a data augmentation method for neural machine translation. It works by interpreting language models and phrasal alignment causally. Specifically, it creates augmented parallel translation corpora by generating (path-specific) counterfactual aligned phrases. We generate these by sampling new source phrases from a masked language model, then sampling an aligned counterfactual target phrase by noting that a translation language model can be interpreted as a Gumbel-Max Structural Causal Model (Oberst and Sontag, 2019). Compared to previous work, our method takes both context and alignment into account to maintain the symmetry between source and target sequences. Experiments on IWSLT{'}15 English → Vietnamese, WMT{'}17 English → German, WMT{'}18 English → Turkish, and WMT{'}19 robust English → French show that the method can improve the performance of translation, backtranslation and translation robustness.

PaperPDFCode

Code

marziehf/DataAugmentationNMT officialmentioned in papertorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Data AugmentationLanguage ModelingLanguage ModellingMachine TranslationTranslation

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections