Training Neural Machine Translation To Apply Terminology Constraints
This addresses the need for efficient and robust terminology application in machine translation, particularly for domain-specific or real-time use cases, though it is incremental as it builds on existing training-based approaches.
The paper tackles the problem of incorporating custom terminology into neural machine translation without computational overhead by training the system to learn terminology usage, resulting in a method that is more effective than state-of-the-art constrained decoding and as fast as constraint-free decoding.
This paper proposes a novel method to inject custom terminology into neural machine translation at run time. Previous works have mainly proposed modifications to the decoding algorithm in order to constrain the output to include run-time-provided target terms. While being effective, these constrained decoding methods add, however, significant computational overhead to the inference step, and, as we show in this paper, can be brittle when tested in realistic conditions. In this paper we approach the problem by training a neural MT system to learn how to use custom terminology when provided with the input. Comparative experiments show that our method is not only more effective than a state-of-the-art implementation of constrained decoding, but is also as fast as constraint-free decoding.