I Have an Attention Bridge to Sell You: Generalization Capabilities of Modular Translation Architectures
This addresses the problem of efficient and generalizable machine translation models for NLP researchers, showing that modular designs may not offer advantages as claimed.
The paper investigates whether modular translation architectures, particularly attention bridges, improve translation quality and generalization, finding that non-modular architectures are comparable or preferable under a fixed computational budget.
Modularity is a paradigm of machine translation with the potential of bringing forth models that are large at training time and small during inference. Within this field of study, modular approaches, and in particular attention bridges, have been argued to improve the generalization capabilities of models by fostering language-independent representations. In the present paper, we study whether modularity affects translation quality; as well as how well modular architectures generalize across different evaluation scenarios. For a given computational budget, we find non-modular architectures to be always comparable or preferable to all modular designs we study.