Controlling Target Features in Neural Machine Translation via Prefix Constraints

We propose prefix constraints, a novel method to enforce constraints on target sentences in neural machine translation. It places a sequence of special tokens at the beginning of target sentence (target prefix), while side constraints (Sennrich et al., 2016) places a special token at the end of source sentence (source suffix). Prefix constraints can be predicted from source sentence jointly with target sentence, while side constraints must be provided by the user or predicted by some other methods. In both methods, special tokens are designed to encode arbitrary features on target-side or metatextual information. We show that prefix constraints are more flexible than side constraints and can be used to control the behavior of neural machine translation, in terms of output length, bidirectional decoding, domain adaptation, and unaligned target word generation.