Simultaneous Translation with Flexible Policy via Restricted Imitation Learning

39383

Knowledge Triples