Prompting Neural Machine Translation with Translation Memories
This addresses the need for efficient and accessible translation memory integration for machine translation practitioners, offering a simple, incremental update to existing systems.
The paper tackled the problem of integrating translation memories into neural machine translation systems without requiring architectural changes or retraining, by treating them as prompts at test time, resulting in significant performance improvements over strong baselines on several datasets.
Improving machine translation (MT) systems with translation memories (TMs) is of great interest to practitioners in the MT community. However, previous approaches require either a significant update of the model architecture and/or additional training efforts to make the models well-behaved when TMs are taken as additional input. In this paper, we present a simple but effective method to introduce TMs into neural machine translation (NMT) systems. Specifically, we treat TMs as prompts to the NMT model at test time, but leave the training process unchanged. The result is a slight update of an existing NMT system, which can be implemented in a few hours by anyone who is familiar with NMT. Experimental results on several datasets demonstrate that our system significantly outperforms strong baselines.