| tech,30-1-H01-1042,bq | output </term> of <term> machine translation ( | MT | ) systems </term> . We believe that these | #577 The purpose of this research is to test the efficacy of applying automated evaluation techniques, originally devised for the evaluation of human language learners, to the output of machine translation ( MT) systems. |
| other,16-3-H01-1042,bq | the <term> intelligibility </term> of <term> | MT | output </term> . A <term> language learning | #625 This, the first experiment in a series of experiments, looks at the intelligibility ofMT output. |
| tech,22-3-N04-1022,bq | word-to-word alignments </term> from an <term> | MT | system </term> , and <term> syntactic structure | #6597 We describe a hierarchy of loss functions that incorporate different levels of linguistic information from word strings, word-to-word alignments from anMT system, and syntactic structure from parse-trees of source and target language sentences. |
| tech,11-5-N04-1022,bq | </term> can be used to tune <term> statistical | MT | </term> performance for specific <term> loss | #6638 Our results show that MBR decoding can be used to tune statistical MT performance for specific loss functions. |
| tech,4-1-I05-2014,bq | metrics </term> for <term> Machine Translation ( | MT | ) systems </term> , such as <term> BLEU </term> | #7685 Automatic evaluation metrics for Machine Translation ( MT) systems, such as BLEU or NIST, are now well established. |
| tech,31-4-I05-2014,bq | </term> with , for instance , <term> statistical | MT | systems </term> which usually segment their | #7778 The use of BLEU at the character level eliminates the word segmentation problem: it makes it possible to directly compare commercial systems outputting unsegmented texts with, for instance, statistical MT systems which usually segment their outputs. |
| other,5-4-I05-4010,bq | particularly valuable to <term> empirical | MT | research </term> . This piece of work has | #8290 It is particularly valuable to empirical MT research. |
| measure(ment),3-1-I05-5003,bq | . The task of <term> machine translation ( | MT | ) evaluation </term> is closely related to | #8322 The task of machine translation ( MT) evaluation is closely related to the task of sentence-level semantic equivalence classification. |
| measure(ment),8-2-I05-5003,bq | investigates the utility of applying standard <term> | MT | evaluation methods ( BLEU , NIST , WER | #8345 This paper investigates the utility of applying standardMT evaluation methods (BLEU, NIST, WER and PER) to building classifiers to predict semantic equivalence and entailment. |
| measure(ment),4-4-I05-5003,bq | sentence </term> . Our results show that <term> | MT | evaluation techniques </term> are able to | #8401 Our results show thatMT evaluation techniques are able to produce useful features for paraphrase classification and to a lesser extent entailment. |
| measure(ment),17-3-I05-5008,bq | reference sets </term> to be used for <term> | MT | evaluation </term> . This paper proposes | #8562 The paraphrase sets produced by this method thus seem adequate as reference sets to be used forMT evaluation. |
| tech,6-5-J05-4003,bq | . We also show that a good-quality <term> | MT | system </term> can be built from scratch | #9074 We also show that a good-qualityMT system can be built from scratch by starting with a very small parallel corpus (100,000 words) and exploiting a large non-parallel corpus. |
| tool,15-3-P05-1048,bq | </term> for a typical <term> IBM statistical | MT | system </term> , we find that <term> word sense | #9364 Using a state-of-the-art Chinese word sense disambiguation model to choose translation candidates for a typical IBM statistical MT system, we find that word sense disambiguation does not yield significantly better translation quality than the statistical machine translation system alone. |
| other,16-4-P05-1048,bq | limitations of current <term> statistical | MT | architectures </term> . <term> Syntax-based | #9405 Error analysis suggests several key factors behind this surprising finding, including inherent limitations of current statistical MT architectures. |
| tech,0-1-P05-1067,bq | Syntax-based statistical machine translation ( | MT | ) </term> aims at applying <term> statistical | #9413 Syntax-based statistical machine translation ( MT) aims at applying statistical models to structured data. |
| tech,6-7-P05-1067,bq | evaluate the <term> outputs </term> of our <term> | MT | system </term> using the <term> NIST and Bleu | #9515 We evaluate the outputs of ourMT system using the NIST and Bleu automatic MT evaluation software. |
| measure(ment),10-7-P05-1067,bq | </term> using the <term> NIST and Bleu automatic | MT | evaluation software </term> . The result | #9523 We evaluate the outputs of our MT system using the NIST and Bleu automatic MT evaluation software. |
| tech,10-2-P05-3025,bq | syntax-based statistical machine translation ( | MT | ) </term> , to understand the <term> model </term> | #9863 The method allows a user to explore a model of syntax-based statistical machine translation ( MT), to understand the model's strengths and weaknesses, and to compare it to other MT systems. |
| tech,33-2-P05-3025,bq | weaknesses , and to compare it to other <term> | MT | systems </term> . Using this <term> visualization | #9881 The method allows a user to explore a model of syntax-based statistical machine translation (MT), to understand the model's strengths and weaknesses, and to compare it to otherMT systems. |
| tech,16-3-P05-3025,bq | conceptual and practical problems in an <term> | MT | system </term> . In our demonstration at <term> | #9900 Using this visualization method, we can find and address conceptual and practical problems in anMT system. |
