Interactive Visualization and Manipulation of Attention-based Neural Machine Translation
2017-09-01EMNLP 2017Unverified0· sign in to hype
Jaesong Lee, Joong-Hwi Shin, Jun-Seok Kim
Unverified — Be the first to reproduce this paper.
ReproduceAbstract
While neural machine translation (NMT) provides high-quality translation, it is still hard to interpret and analyze its behavior. We present an interactive interface for visualizing and intervening behavior of NMT, specifically concentrating on the behavior of beam search mechanism and attention component. The tool (1) visualizes search tree and attention and (2) provides interface to adjust search tree and attention weight (manually or automatically) at real-time. We show the tool gives various methods to understand NMT.