Viterbi Decoding of Directed Acyclic Transformer for Non-Autoregressive Machine Translation
This work addresses the trade-off between speed and accuracy in machine translation for NLP applications, representing an incremental improvement over existing methods.
The paper tackled the problem of sequential dependency in non-autoregressive machine translation by proposing a Viterbi decoding framework for Directed Acyclic Transformer, which improved translation accuracy while maintaining fast decoding speeds, as shown in experimental results.
Non-autoregressive models achieve significant decoding speedup in neural machine translation but lack the ability to capture sequential dependency. Directed Acyclic Transformer (DA-Transformer) was recently proposed to model sequential dependency with a directed acyclic graph. Consequently, it has to apply a sequential decision process at inference time, which harms the global translation accuracy. In this paper, we present a Viterbi decoding framework for DA-Transformer, which guarantees to find the joint optimal solution for the translation and decoding path under any length constraint. Experimental results demonstrate that our approach consistently improves the performance of DA-Transformer while maintaining a similar decoding speedup.