Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Lexically Constrained Decoding for Sequence Generation Using Grid Beam Search

About

We present Grid Beam Search (GBS), an algorithm which extends beam search to allow the inclusion of pre-specified lexical constraints. The algorithm can be used with any model that generates a sequence $ \mathbf{\hat{y}} = \{y_{0}\ldots y_{T}\} $, by maximizing $ p(\mathbf{y} | \mathbf{x}) = \prod\limits_{t}p(y_{t} | \mathbf{x}; \{y_{0} \ldots y_{t-1}\}) $. Lexical constraints take the form of phrases or words that must be present in the output sequence. This is a very general way to incorporate additional knowledge into a model's output without requiring any modification of the model parameters or training data. We demonstrate the feasibility and flexibility of Lexically Constrained Decoding by conducting experiments on Neural Interactive-Predictive Translation, as well as Domain Adaptation for Neural Machine Translation. Experiments show that GBS can provide large improvements in translation quality in interactive scenarios, and that, even without any user input, GBS can be used to achieve significant gains in performance in domain adaptation scenarios.

Chris Hokamp, Qun Liu• 2017

Related benchmarks

TaskDatasetResultRank
Table-to-text generationE2ENLG (test)
BLEU40.7
37
Commonsense GenerationCommonGen (test)
ROUGE-L38.2
31
Showing 2 of 2 rows

Other info

Follow for update