Evaluating Syntactic Properties of Seq2seq Output with a Broad Coverage HPSG: A Case Study on Machine Translation
This addresses the problem of understanding syntactic quality in neural machine translation outputs for researchers and practitioners, though it is incremental as it applies an existing grammar analysis to new model data.
The study assessed whether seq2seq machine translation models produce syntactically valid English by using the English Resource Grammar, finding that over 93% of translations were parseable, but the model struggled with rarer syntactic rules and specific constructions compared to references.
Sequence to sequence (seq2seq) models are often employed in settings where the target output is natural language. However, the syntactic properties of the language generated from these models are not well understood. We explore whether such output belongs to a formal and realistic grammar, by employing the English Resource Grammar (ERG), a broad coverage, linguistically precise HPSG-based grammar of English. From a French to English parallel corpus, we analyze the parseability and grammatical constructions occurring in output from a seq2seq translation model. Over 93\% of the model translations are parseable, suggesting that it learns to generate conforming to a grammar. The model has trouble learning the distribution of rarer syntactic rules, and we pinpoint several constructions that differentiate translations between the references and our model.