This article presents methods for biasing statistical translation models to reflect these properties. Evaluation with respect to independent human judgments has confirmed that translation models biased in this fashion are significantly more accurate than a baseline knowledge-free model. This article also shows how a statistical translation model can take advantage of preexisting knowledge that might be available about particular language pairs. Even the simplest kinds of languagespecific knowledge, such as the distinction between content words and function words, are shown to reliably boost translation model performance on some tasks. Statistical models that reflect knowledge about the model domain combine the best of both the rationalist and empiricist paradigms

Abstract not found

Multitext Grammars (MTGs) generate arbitrarily many parallel texts via production rules of arbitrary length. Both ordinary MTGs and their bilexical subclass admit relatively efficient parsers. Yet, MTGs are more expressive than other synchronous formalisms for which parsers have been described in the literature. The combination of greater expressive power and relatively low cost of inference makes MTGs an attractive foundation for practical models of translational equivalence.

The formalism of synchronous tree-adjoining grammars, a variant of standard tree-adjoining grammars (TAG), was intended to allow the use of TAGs for language transduction in addition to language specification. In previous work, the definition of the transduction relation defined by a synchronous TAG was given by appeal to an iterative rewriting process. The rewriting definition of derivation is problematic in that it greatly extends the expressivity of the formalism and makes the design of parsing algorithms difficult if not impossible. We introduce a simple, natural definition of synchronous tree-adjoining derivation, based on isomorphisms between standard tree-adjoining derivations, that avoids the expressivity and implementability problems of the original rewriting definition. The decrease in expressivity, which would otherwise make the method unusable, is offset by the incorporation of an alternative definition of standard tree-adjoining derivation, previously proposed for complete...

Research in computational linguistics, computer graphics and autonomous agents has led to the development of increasingly sophisticated communicative agents over the past few years, bringing new perspective to machine translation research. The engineering of language-based smooth, expressive, natural-looking human gestures can give us useful insights into the design principles that have evolved in natural communication between people. In this paper we prototype a machine translation system from English to American Sign Language (ASL), taking into account not only linguistic but also visual and spatial information associated with ASL signs.

. In this paper we present one aspect of our research on machine translation (MT): capturing the grammatical and computational relation between (i) the interlingua (IL) as defined declaratively in the lexicon and (ii) the IL as defined procedurally by way of algorithms that compose and decompose pivot IL forms. We begin by examining the interlinguas in the lexicons of a variety of current IL-based approaches to MT. This brief survey makes it clear that no consensus exists among MT researchers on the level of representation for defining the IL. In the section that follows, we explore the consequences of this missing formal framework for MT system builders who develop their own lexical-IL entries. The lack of software tools to support rapid IL respecification and testing greatly hampers their ability to modify representations to handle new data and new domains. Our view is that IL-based MT research needs both (a) the formal framework to specify possible IL grammars and (b) the software ...

this paper introduces a new formalism, the dynamic link synchronous tree--adjoining grammars or DLS--TAGs.

this paper. At a higher level of description, the grammar rules and principles that are implicit in the form of the lexicon would be stated explicitly. For example, there are principles which govern which trees are grouped together into tree families, and rules which describe the relations between structure types across tree families (see subsection 2.1 for a discussion of tree families.) The information explicitly provided in this more abstract representation of the grammar may be thought of as an interpretation of the data in the lower-level representation.

12.1> ff 1 * S NP John V P V misses NP 1 # S NP 0 # V P V manque PP P `a NP John + Then, if the pair fi operates on the NP 1 -NP 0 in ff 1 , the following pair ff 2 is generated. ff 2 * S NP John V P V misses NP Mary S NP Mary V P V manque PP P `a NP John + 1 ff * NP John NP John + fi * NP Mary NP Mary + / * S NP 0 # V P V misses NP 1 # S NP 0 # V P V manque PP P `a NP 1 # + ffi * S Adv apparently S S Adv apparement S +<F12.

iv guages is determined only in part by the synchronization. The base formalism chosen can offer greater or lesser opportunity for divergence in the derived structures. My choice of a base formalism is motivated directly by research into applications of synchronous TAG-based grammars to two natural language applications: semantic interpretation and natural language translations. I first examine a range of TAG variants that have not previously been studied in this level of detail to determine their computational properties and to develop algorithms that can be used to process them. Original results on the complexity of these formalisms are presented as well as novel algorithms for factorizing grammars to reduce the time required to process them. In Part II, I develop applications of synchronous Limited Delay Tree-Local Multicomponent TAG to semantic interpretation and probabilistic synchronous Tree Insertion Grammar to statistical natural language translation. Contents