In this paper, we address the question, ’~/hat types of MT divergences and mismatches must be accommodated in the lexicon?" In our work, we have focused on the treatment of divergences and mismatches in an interlingual (as opposed to a tranfer-based) MT system. In such systems, one uses monolingual lexicons for each of the language under consideration, so divergences and mismatches are not hand...

We present a prototype interlingual communication system that is being used to collect a corpus of task based dialogues between speakers of different languages. This corpus will be used to assess human reactions to an automated speech-to-speech translation system. In this demonstration we show how the HCRC Map Task can be adapted to support data collection in this interlingual environment, and ...

Many projects are conducted to develop multilingual lexical databases. Some of these projects use an interlingual approach (KBMT-89, EDR, ...), where others choose a bilingual approach (Multilex, ...). This paper presents an interlingual approach based on acceptions (word-senses) aiming at the development of a multilingual lexical database management system: NADIA. With this approach, the inter...

Machine translation has been a particularly difficult problem in the area of Natural Language Processing for over two decades. Early approaches to translation failed in part because interaction effects of complex phenomena made translation appear to be unmanageable. Later approaches to the problem have succeeded but are based on many language-specific rules. To capture all natural language phen...

The first part of this paper considers some of the reasons why mechanical translation via a logically formalized interlingua is worth pursuing. The interlingua described consists of a network of bonded semantic elements, the bonds being either homogeneous, corresponding to a generalized notion of qualification, or heterogeneous, for dyadic relations. The translation procedure involves a basic p...

This paper describes a multi-level design, i.e., a non-uniform approach to interlingual machine translation (MT), in which distinct representational languages are used for diierent types of knowledge. We demonstrate that a linguistically-motivated \division of labor" across multiple representation levels has not complicated, but rather has readily facilitated, the identiication and construction...

This paper presents a method for creating interlingual wordto-word or phrase-to-phrase mappings between any two languages using the self-organizing map algorithm. The method can be used as a component in a statistical machine translation system. The conceptual space created by the self-organizing map serves as a kind of interlingual representation. The specific problems of machine translation a...

The IAMTC project (Interlingual Annotation of Multilingual Translation Corpora) is developing an interlingual representation framework for annotation of parallel corpora (English paired with Arabic, French, Hindi, Japanese, Korean, and Spanish) with deep-semantic representations. In particular, we are investigating meaning equivalent paraphrases involving conversives and non-literal language us...

Dissertation title : Discriminative Interlingual Representations Jagadeesh Jagarlamudi, Doctor of Philosophy, 2013 Dissertation advised by: Hal Daumé III Department of Computer Science The language barrier in many multilingual natural language processing (NLP) tasks can be overcome bymapping objects from different languages (“views”) into a common low-dimensional subspace. For example, the name...