Results 1 
9 of
9
Better kbest parsing
, 2005
"... We discuss the relevance of kbest parsing to recent applications in natural language processing, and develop efficient algorithms for kbest trees in the framework of hypergraph parsing. To demonstrate the efficiency, scalability and accuracy of these algorithms, we present experiments on Bikel’s i ..."
Abstract

Cited by 185 (17 self)
 Add to MetaCart
We discuss the relevance of kbest parsing to recent applications in natural language processing, and develop efficient algorithms for kbest trees in the framework of hypergraph parsing. To demonstrate the efficiency, scalability and accuracy of these algorithms, we present experiments on Bikel’s implementation of Collins ’ lexicalized PCFG model, and on Chiang’s CFGbased decoder for hierarchical phrasebased translation. We show in particular how the improved output of our algorithms has the potential to improve results from parse reranking systems and other applications. 1
Fast Decoding and Optimal Decoding for Machine Translation
 In Proceedings of ACL 39
, 2001
"... A good decoding algorithm is critical ..."
(Show Context)
Building a Statistical Machine Translation System from Scratch: How Much Bang for the Buck Can We Expect?
, 2001
"... Introduction Crises and disasters frequently attract international attention to regions of the world that have previously been largely ignored by the international community. While it is possible to stock up on emergency relief supplies and, for the worst case, weapons, regardless of where exactly ..."
Abstract

Cited by 11 (2 self)
 Add to MetaCart
(Show Context)
Introduction Crises and disasters frequently attract international attention to regions of the world that have previously been largely ignored by the international community. While it is possible to stock up on emergency relief supplies and, for the worst case, weapons, regardless of where exactly they are eventually going to be used, this cannot be done with multilingual information processing technology. This technology will often have to be developed after the fact in a quick response to the given situation. Multilingual data resources for statistical approaches, such as parallel corpora, may not always be available. In the fall of 2000, we decided to put the current state of the art to the test with respect to the rapid construction of a machine translation system from scratch. Within one month, we would # hire translators; # translate as much text as possible; and # train a statistical MT system on the data thus created. The languag
Fast and Optimal Decoding for Machine Translation
 ARTIFICIAL INTELLIGENCE
, 2004
"... A good decoding algorithm is critical to the success of any statistical machine translation system. The decoder's job is to find the translation that is most likely according to set of previously learned parameters (and a formula for combining them). Since the space of possible translations is ..."
Abstract

Cited by 7 (0 self)
 Add to MetaCart
A good decoding algorithm is critical to the success of any statistical machine translation system. The decoder's job is to find the translation that is most likely according to set of previously learned parameters (and a formula for combining them). Since the space of possible translations is extremely large, typical decoding algorithms are only able to examine a portion of it, thus risking to miss good solutions. Unfortunately, examining more of the space leads to unacceptably slow decodings. In this
Building a Statistical Machine Translation System from Scratch:
"... We report on our experience with building a statistical MT system from scratch, including the creation of a small parallel TamilEnglish corpus, and the results of a taskbased pilot evaluation of statistical MT systems trained on sets of ca. 1300 and ca. ..."
Abstract
 Add to MetaCart
We report on our experience with building a statistical MT system from scratch, including the creation of a small parallel TamilEnglish corpus, and the results of a taskbased pilot evaluation of statistical MT systems trained on sets of ca. 1300 and ca.
Statistical Machine Translation of French and German into English Using IBM Model 2 Greedy Decoding
"... The job of a decoder in statistical machine translation is to find the most probable translation of a given sentence, as defined by a set of previously learned parameters. Because the search space of potential translations is essentially infinite, there is always a tradeoff between accuracy and sp ..."
Abstract
 Add to MetaCart
(Show Context)
The job of a decoder in statistical machine translation is to find the most probable translation of a given sentence, as defined by a set of previously learned parameters. Because the search space of potential translations is essentially infinite, there is always a tradeoff between accuracy and speed when designing a decoder. Germann et al. [4] recently presented a fast, greedy decoder that starts with an initial guess and then refines that guess through small “mutations ” that produce more probable translations. The greedy decoder in [4] was designed to work with the IBM Model 4 translation model, which, while being a sophisticated model of the translation process, is also quite complex and therefore difficult to implement and fairly slow in training and decoding. We present modifications to the greedy decoder presented in [4] that allow it to work with the simpler and more efficient IBM Model 2. We have tested our modified decoder by having it translate equivalent French and German sentences into English, and we present the results and translation accuracies that we have obtained. Because we are interested in the relative effectiveness of our decoder in translating between different languages, we discuss the discrepancies between the results we obtained when performing FrenchtoEnglish and GermantoEnglish translation, and we speculate on the factors inherent to these languages that may have contributed to these discrepancies. 1. Introduction and Related
unknown title
"... We discuss the relevance of kbest parsing to recent applications in natural language processing, and develop efficient algorithms for kbest trees in the framework of hypergraph parsing. To demonstrate the efficiency, scalability and accuracy of these algorithms, we present experiments on Bikel’s i ..."
Abstract
 Add to MetaCart
(Show Context)
We discuss the relevance of kbest parsing to recent applications in natural language processing, and develop efficient algorithms for kbest trees in the framework of hypergraph parsing. To demonstrate the efficiency, scalability and accuracy of these algorithms, we present experiments on Bikel’s implementation of Collins ’ lexicalized PCFG model, and on Chiang’s CFGbased decoder for hierarchical phrasebased translation. We show in particular how the improved output of our algorithms has the potential to improve results from parse reranking systems and other applications. 1
Computing Optimal Alignments for the IBM3 Translation Model
"... Prior work on training the IBM3 translation model is based on suboptimal methods for computing Viterbi alignments. In this paper, we present the first method guaranteed to produce globally optimal alignments. This not only results in improved alignments, it also gives us the opportunity to evaluate ..."
Abstract
 Add to MetaCart
(Show Context)
Prior work on training the IBM3 translation model is based on suboptimal methods for computing Viterbi alignments. In this paper, we present the first method guaranteed to produce globally optimal alignments. This not only results in improved alignments, it also gives us the opportunity to evaluate the quality of standard hillclimbing methods. Indeed, hillclimbing works reasonably well in practice but still fails to find the global optimum for between 2 % and 12 % of all sentence pairs and the probabilities can be several tens of orders of magnitude away from the Viterbi alignment. By reformulating the alignment problem as an Integer Linear Program, we can use standard machinery from global optimization theory to compute the solutions. We use the wellknown branchandcut method, but also show how it can be customized to the specific problem discussed in this paper. In fact, a large number of alignments can be excluded from the start without losing global optimality. 1