Errors in English parse trees impact the quality of syntax-based MT systems trained using those parses. Frequent sources of error for English parsers include PP-attachment ambiguity, NP-bracketing ambiguity, and coordination ambiguity. Not all ambiguities are preserved across languages. We examine a common type of ambiguity in English that is not preserved in Chinese: given a sequence “VP NP PP...

This paper examines an anaphoric construction, British English do, and locates it within the dichotomy in the ellipsis literature between deleted phrases and null pro-forms, concluding that the choice is a false one, in that pro-forms involve deletion as well; the question, then, is how to account for the differential permeability to dependencies that require external licensing of the various d...

The purpose of this article is to characterise the production of postverbal subjects in two ICLE subcorpora (Italian and Spanish). The question has been dealt with before in the literature with emphasis on the production of ungrammatical inversion structures in L2 English of speakers from a variety of L1s, but in quite a scattered, unsystematic and rather intuitive fashion. Our approach seeks t...

This thesis challenges the claim that syntactic change should be modeled in terms of adjustments in the probability distribution of competing grammars (Kroch 1989b; Yang 2000). Building onwork of Boersma (1998), Boersma and Hayes (2001), and Bresnan et al. (2001), I propose a novel formal framework for modeling syntactic change that is firmly embedded in the inherent variability tradition (Wein...

The general problem area of machine translation has been outlined and related to other fields of language-data processing in Garvin's “A Linguist's View of Language-data Processing.” It has been discussed in some detail in the two immediately preceding sections by Hays and Harper. This discussion will deal more specifically with the question of syntax: both with the place of syntax in the over-...

We present a new unsupervised syntax-based MT system, termed U-DOT, which uses the unsupervised U-DOP model for learning paired trees, and which computes the most probable target sentence from the relative frequencies of paired subtrees. We test U-DOT on the German-English Europarl corpus, showing that it outperforms the state-of-the-art phrase-based Pharaoh system. We demonstrate that the incl...

In this paper, we give an overview of the ICT statistical machine translation systems for the evaluation campaign of the International Workshop on Spoken Language Translation (IWSLT) 2007. In this year’s evaluation, we participated in the Chinese-English transcript translation task, and developed three systems based on different techniques: a formally syntax-based system Bruin, an extended phra...

This paper reports recent progress on modeling the grounded co-acquisition of syntax and semantics of locative spatial language in developmental robots. We show how a learner robot can learn to produce and interpret spatial utterances in guided-learning interactions with a tutor robot (equipped with a system for producing English spatial phrases). The tutor guides the learning process by simpli...