TL;DR研究了使用 Wait-k 解码器在资源匮乏的情况下进行同声传译的行为,同时通过使用单向编码器和训练多个 K 值来提高模型的训练,并探讨了 2D 卷积体系结构在同声翻译方面的应用。
Abstract
simultaneous machine translation consists in starting output generation
before the entire input sequence is available. wait-k decoders offer a simple
but efficient approach for this problem. They first read k sou