TEXAN is a sysLem of bransfer, .or iented t ex t ana lys is . I ts l i n g u i s t i c concept is based on a communicative appY'oacll w i th in the framework of speech act theory. In th is view tex ts are considered to be the resu l t of l i n g u i s t i c act ions° I t is assumed tha t they contro l the se lect ion of t r a n s l a t i o n equ iva lents . The t ransi t i o n of th i s concept...

As Machine Translation (MT) becomes more popular among end-users, an increasingly relevant issue is that of estimating the quality of automatic translations for a particular task. The main application for such quality estimates has been selecting good enough translations for human post-editing. The endusers, in this case, are fluent speakers of both source and target languages and the quality e...

H a v i n g a g r e e d , w i t h g r a t i t u d e and d e l i g h t , t o s e r v e a s c h a i r m a n o f t h e m a c h i n e t r a n s l a t i o n s e s s i o n o f t h i s C o n f e r e n c e , i t became my d u t y t o comment f o r t h e r e c o r d on t h e s e t o f f o u r p a p e r s t o be p r e s e n t e d t o u s . Thus h a v e e v e n t s l a i d on me an o p p o r t u n i t y t...

We propose a Name-aware Machine Translation (MT) approach which can tightly integrate name processing into MT model, by jointly annotating parallel corpora, extracting name-aware translation grammar and rules, adding name phrase table and name translation driven decoding. Additionally, we also propose a new MT metric to appropriately evaluate the translation quality of informative words, by ass...

Logos analyzes whole source sentences, considering morphology, meaning, and grammatical structure and function. The analysis determines the semantic relationships between words as well as the syntactic structure of the sentence. Parsing is only source language-specific; generation is target language-specific. Although the Logos system was originally developed on the basis of a transfer approach...

We present a new approach for neural machine translation (NMT) using the morphological and grammatical decomposition of the words (factors) in the output side of the neural network. This architecture addresses two main problems occurring in MT, namely dealing with a large target language vocabulary and the out of vocabulary (OOV) words. By the means of factors, we are able to handle larger voca...

A fully transistorized computer for the purpose of experimental machine translation from English to Japanese was recently completed, and a program test is now being conducted. This paper describes the organization of this special purpose computer named Yamato as well as the translation principles and the program flow diagrams. Yamato, whose switching elements use 600 transistors and 7,000 germa...

Many of the problems with (and consequently much of the resistance to) machine translation (MT) has sprung from ignorance: on the part of early developers, who failed to appreciate the magnitude of the problems they faced, and hence made outrageous claims; on the part of opponents, who frequently fail to understand what translation is about, and how it is (properly) performed; on the part of ze...

We describe sister machine translation prototypes, Ntran, an English to Japanese system developed at UMIST, and Aidtrans, Japanese to English, at Sheffield, both designed for use by an English monolingual. Aidtraus uses extensive and sophisticated collocational analysis radically to reduce the need for conventional post-editing. Ntran offers interactive query at three stages: on-line dictionary...

In this paper, we study a new learning paradigm for Neural Machine Translation (NMT). Instead of maximizing the likelihood of the human translation as in previous works, we minimize the distinction between human translation and the translation given by a NMT model. To achieve this goal, inspired by the recent success of generative adversarial networks (GANs), we employ an adversarial training a...