Towards Understanding Neural Machine Translation with Word Importance
This work addresses interpretability for NMT researchers and practitioners, but it is incremental as it builds on existing gradient-based attribution methods.
The paper tackled the problem of poor interpretability in neural machine translation by measuring word importance using a gradient-based method to attribute outputs to inputs, showing it can identify influential words and under-translated ones, with validation across language pairs and models.
Although neural machine translation (NMT) has advanced the state-of-the-art on various language pairs, the interpretability of NMT remains unsatisfactory. In this work, we propose to address this gap by focusing on understanding the input-output behavior of NMT models. Specifically, we measure the word importance by attributing the NMT output to every input word through a gradient-based method. We validate the approach on a couple of perturbation operations, language pairs, and model architectures, demonstrating its superiority on identifying input words with higher influence on translation performance. Encouragingly, the calculated importance can serve as indicators of input words that are under-translated by NMT models. Furthermore, our analysis reveals that words of certain syntactic categories have higher importance while the categories vary across language pairs, which can inspire better design principles of NMT architectures for multi-lingual translation.