Classical Structured Prediction Losses for Sequence to Sequence Learning

There has been much recent work on training neural attention models at the sequence-level using either reinforcement learning-style methods or by optimizing the beam. In this paper, we survey a range of classical objective functions that have been widely used to train linear models for structured prediction and apply them to neural sequence to sequence models... Our experiments show that these losses can perform surprisingly well by slightly outperforming beam search optimization in a like for like setup. We also report new state of the art results on both IWSLT'14 German-English translation as well as Gigaword abstractive summarization. On the larger WMT'14 English-French translation task, sequence-level training achieves 41.5 BLEU which is on par with the state of the art. read more

PDF Abstract NAACL 2018 PDF NAACL 2018 Abstract

Datasets


  Add Datasets introduced or used in this paper
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Machine Translation IWSLT2014 German-English Minimum Risk Training [Edunov2017] BLEU score 32.84 # 23
Machine Translation IWSLT2015 German-English ConvS2S+Risk BLEU score 32.93 # 4

Methods


No methods listed for this paper. Add relevant methods here