English

A Streaming Approach For Efficient Batched Beam Search

Computation and Language 2021-08-17 v3 Artificial Intelligence Distributed, Parallel, and Cluster Computing Machine Learning Performance

Abstract

We propose an efficient batching strategy for variable-length decoding on GPU architectures. During decoding, when candidates terminate or are pruned according to heuristics, our streaming approach periodically "refills" the batch before proceeding with a selected subset of candidates. We apply our method to variable-width beam search on a state-of-the-art machine translation model. Our method decreases runtime by up to 71% compared to a fixed-width beam search baseline and 17% compared to a variable-width baseline, while matching baselines' BLEU. Finally, experiments show that our method can speed up decoding in other domains, such as semantic and syntactic parsing.

Keywords

Cite

@article{arxiv.2010.02164,
  title  = {A Streaming Approach For Efficient Batched Beam Search},
  author = {Kevin Yang and Violet Yao and John DeNero and Dan Klein},
  journal= {arXiv preprint arXiv:2010.02164},
  year   = {2021}
}

Comments

EMNLP 2020

R2 v1 2026-06-23T19:03:14.820Z