Later-stage Minimum Bayes-Risk Decoding for Neural Machine Translation

11 Apr 2017  ·  Raphael Shu, Hideki Nakayama ·

For extended periods of time, sequence generation models rely on beam search algorithm to generate output sequence. However, the correctness of beam search degrades when the a model is over-confident about a suboptimal prediction. In this paper, we propose to perform minimum Bayes-risk (MBR) decoding for some extra steps at a later stage. In order to speed up MBR decoding, we compute the Bayes risks on GPU in batch mode. In our experiments, we found that MBR reranking works with a large beam size. Later-stage MBR decoding is shown to outperform simple MBR reranking in machine translation tasks.

PDF Abstract


  Add Datasets introduced or used in this paper

Results from the Paper

  Submit results from this paper to get state-of-the-art GitHub badges and help the community compare results to other papers.