This work extends phrase-based statistical MT (SMT) with shallow syntax dependencies. Two string-to-chunks translation models are proposed: a factored model, which augments phrase-based SMT with layered dependencies, and a joint model, that extends the phrase translation table with microtags, i.e. perword projections of chunk labels. Both rely on n-gram models of target sequences with different...

This paper describes a method using commonsense knowledge for discarding spurious syntactic ambiguities introduced by postverbal prepositional phrase attachment during parsing. A completely naive parser will generate three parses for sentences of the form NP-V-Det-N-PP. The prepositions alone are insufficiently precise in meaning to guide selection among competing parses. The method is imbedded...

Neural machine translation (NMT) models are able to partially learn syntactic information from sequential lexical information. Still, some complex syntactic phenomena such as prepositional phrase attachment are poorly modeled. This work aims to answer two questions: 1) Does explicitly modeling target language syntax help NMT? 2) Is tight integration of words and syntax better than multitask tra...

Two experiments tested whether phonological phrase boundaries constrain online syntactic analysis in French. Pairs of homophones belonging to different syntactic categories (verb and adjective) were used to create sentences with a local syntactic ambiguity (e.g., [le petit chien mort], in English, the dead little dog, vs. [le petit chien] [mord], in English, the little dog bites, where brackets...

Neural machine translation (NMT) models are able to partially learn syntactic information from sequential lexical information. Still, some complex syntactic phenomena such as prepositional phrase attachment are poorly modeled. This work aims to answer two questions: 1) Does explicitly modeling target language syntax help NMT? 2) Is tight integration of words and syntax better than multitask tra...

This paper takes a closer look at the attraction properties of T. It highlights an empirically attested distinction between rich agreement inflection, exhibited by null-subject languages, and rich tense inflection, found in Romance, but not Germanic, and argues for the syntactic relevance of this distinction. We propose a novel typology of the ways in which T's featural requirements can be sati...

Specific Language Impairment (SLI) is a developmental language disorder characterised by morpho-syntactic errors in the absence of neurological trauma, cognitive impairment, psycho-emotional disturbance, or motor-articulatory disorders. Among the hypotheses that have argued for SLI as an impairment at the linguistic level, rather than at the cognitive or perceptual level, there is a debate as t...