Translate Smart, not Hard: Cascaded Translation Systems with Quality-Aware Deferral
This provides a practical solution for efficient machine translation deployment, though it is incremental as it applies existing methods in a new configuration.
The paper tackles the computational cost of large machine translation models by proposing a cascaded system that uses quality estimation metrics to defer only 30% to 50% of examples to larger models, matching their performance while reducing costs.
Larger models often outperform smaller ones but come with high computational costs. Cascading offers a potential solution. By default, it uses smaller models and defers only some instances to larger, more powerful models. However, designing effective deferral rules remains a challenge. In this paper, we propose a simple yet effective approach for machine translation, using existing quality estimation (QE) metrics as deferral rules. We show that QE-based deferral allows a cascaded system to match the performance of a larger model while invoking it for a small fraction (30% to 50%) of the examples, significantly reducing computational costs. We validate this approach through both automatic and human evaluation.