Non-Monotonic Attention-based Read/Write Policy Learning for Simultaneous Translation

Open in new window