Lexicalized reordering models are very important components of phrasebased translation systems.By examining the reordering relationships between adjacent phrases,conventional methods learn these models from the word a...Lexicalized reordering models are very important components of phrasebased translation systems.By examining the reordering relationships between adjacent phrases,conventional methods learn these models from the word aligned bilingual corpus,while ignoring the effect of the number of adjacent bilingual phrases.In this paper,we propose a method to take the number of adjacent phrases into account for better estimation of reordering models.Instead of just checking whether there is one phrase adjacent to a given phrase,our method firstly uses a compact structure named reordering graph to represent all phrase segmentations of a parallel sentence,then the effect of the adjacent phrase number can be quantified in a forward-backward fashion,and finally incorporated into the estimation of reordering models.Experimental results on the NIST Chinese-English and WMT French-Spanish data sets show that our approach significantly outperforms the baseline method.展开更多
The performance of a machine translation system heavily depends on the quantity and quality of the bilingual language resource. However,getting a parallel corpus,which has a large scale and is of high quality,is a ver...The performance of a machine translation system heavily depends on the quantity and quality of the bilingual language resource. However,getting a parallel corpus,which has a large scale and is of high quality,is a very difficult task especially for low resource languages such as Chinese-Vietnamese. Fortunately,multilingual user generated contents（ UGC）,such as bilingual movie subtitles,provide us access to automatic construction of the parallel corpus. Although the amount of UGC parallel corpora can be considerable,the original corpus is not suitable for statistical machine translation（ SMT） systems. The corpus may contain translation errors,sentence mismatching,free translations,etc. To improve the quality of the bilingual corpus for SMT systems,three filtering methods are proposed： sentence length difference,the semantic of sentence pairs,and machine learning. Experiments are conducted on the Chinese to Vietnamese translation corpus.Experimental results demonstrate that all the three methods effectively improve the corpus quality,and the machine translation performance（ BLEU score） can be improved by 1. 32.展开更多
The technology in modern society is very useful.China has a lot of new advanced technology.Let me introduce some to you.The first one is Artificial Intelligence(AI).China has many AI companies,such as Baidu,Alibaba an...The technology in modern society is very useful.China has a lot of new advanced technology.Let me introduce some to you.The first one is Artificial Intelligence(AI).China has many AI companies,such as Baidu,Alibaba and Tencent.They have made many things such as machine learning,natural language processing and computer vision.展开更多
Over the past decade,artificial intelligence(AI)has evolved at an unprecedented pace,transforming technology,industry,and society.From diagnosing diseases with remarkable accuracy to powering self-driving cars and rev...Over the past decade,artificial intelligence(AI)has evolved at an unprecedented pace,transforming technology,industry,and society.From diagnosing diseases with remarkable accuracy to powering self-driving cars and revolutionizing personalized learning,AI is reshaping our world in ways once thought impossible.Spanning fields such as machine learning,deep learning,natural language processing,robotics,and ChatGPT,AI continues to push the boundaries of innovation.As AI continues to advance,it is vital to have a platform that not only disseminates cutting-edge research innovations but also fosters broad discussions on its societal impact,ethical considerations,and interdisciplinary applications.With this vision in mind,we proudly introduce Artificial Intelligence Science and Engineering(AISE)-a journal dedicated to nurturing the next wave of AI innovation and engineering applications.Our mission is to provide a premier outlet where researchers can share high-quality,impactful studies and collaborate to advance AI across academia,industry,and beyond.展开更多
In this essay,our goal is to discover science in Martin Heidegger’s Introduction to Metaphysics,lecture notes for his 1935 summer semester course,because,after all,his subject is metaphysica generalis,or ontology,and...In this essay,our goal is to discover science in Martin Heidegger’s Introduction to Metaphysics,lecture notes for his 1935 summer semester course,because,after all,his subject is metaphysica generalis,or ontology,and this could be construed as a theory of the human brain.Here,by means of verbatim quotes from his text,we attempt to show that indeed these lectures can be viewed as suggestion for an objective scientific theory of human perception,the human capacity for deciphering phenomena,i.e.hermeneutics in its broadest sense.His added notes from the 1953 edition,all of which are comments,not corrections,imply that he never abandoned these thoughts on metaphysics,despite all of his utterances about a need to overcome it,and their popular interpretations to that effect.In his presentation,he further develops the colorful and intuitive style,an hermeneutic language,that he had created in his earlier work Being and Time.The logical functions of Dasein’s anatomical brain are performed by the logos machine,formerly the human soul,using top-down processing based on a global context,the noumenal cosmos which humans maintain internally.Heidegger’s 1942/43 winter semester lectures Parmenides extend in unbroken fashion his 1935 work,proving that he never abandoned,as is widely claimed,his metaphysical avenue of thought.展开更多
基金supported by the National Natural Science Foundation of China（No.61303082） the Research Fund for the Doctoral Program of Higher Education of China（No.20120121120046）
文摘Lexicalized reordering models are very important components of phrasebased translation systems.By examining the reordering relationships between adjacent phrases,conventional methods learn these models from the word aligned bilingual corpus,while ignoring the effect of the number of adjacent bilingual phrases.In this paper,we propose a method to take the number of adjacent phrases into account for better estimation of reordering models.Instead of just checking whether there is one phrase adjacent to a given phrase,our method firstly uses a compact structure named reordering graph to represent all phrase segmentations of a parallel sentence,then the effect of the adjacent phrase number can be quantified in a forward-backward fashion,and finally incorporated into the estimation of reordering models.Experimental results on the NIST Chinese-English and WMT French-Spanish data sets show that our approach significantly outperforms the baseline method.
基金Supported by the National Basic Research Program of China(973Program)(2013CB329303)the National Natural Science Foundation of China(61502035)
文摘The performance of a machine translation system heavily depends on the quantity and quality of the bilingual language resource. However,getting a parallel corpus,which has a large scale and is of high quality,is a very difficult task especially for low resource languages such as Chinese-Vietnamese. Fortunately,multilingual user generated contents（ UGC）,such as bilingual movie subtitles,provide us access to automatic construction of the parallel corpus. Although the amount of UGC parallel corpora can be considerable,the original corpus is not suitable for statistical machine translation（ SMT） systems. The corpus may contain translation errors,sentence mismatching,free translations,etc. To improve the quality of the bilingual corpus for SMT systems,three filtering methods are proposed： sentence length difference,the semantic of sentence pairs,and machine learning. Experiments are conducted on the Chinese to Vietnamese translation corpus.Experimental results demonstrate that all the three methods effectively improve the corpus quality,and the machine translation performance（ BLEU score） can be improved by 1. 32.
文摘The technology in modern society is very useful.China has a lot of new advanced technology.Let me introduce some to you.The first one is Artificial Intelligence(AI).China has many AI companies,such as Baidu,Alibaba and Tencent.They have made many things such as machine learning,natural language processing and computer vision.
文摘Over the past decade,artificial intelligence(AI)has evolved at an unprecedented pace,transforming technology,industry,and society.From diagnosing diseases with remarkable accuracy to powering self-driving cars and revolutionizing personalized learning,AI is reshaping our world in ways once thought impossible.Spanning fields such as machine learning,deep learning,natural language processing,robotics,and ChatGPT,AI continues to push the boundaries of innovation.As AI continues to advance,it is vital to have a platform that not only disseminates cutting-edge research innovations but also fosters broad discussions on its societal impact,ethical considerations,and interdisciplinary applications.With this vision in mind,we proudly introduce Artificial Intelligence Science and Engineering(AISE)-a journal dedicated to nurturing the next wave of AI innovation and engineering applications.Our mission is to provide a premier outlet where researchers can share high-quality,impactful studies and collaborate to advance AI across academia,industry,and beyond.
文摘In this essay,our goal is to discover science in Martin Heidegger’s Introduction to Metaphysics,lecture notes for his 1935 summer semester course,because,after all,his subject is metaphysica generalis,or ontology,and this could be construed as a theory of the human brain.Here,by means of verbatim quotes from his text,we attempt to show that indeed these lectures can be viewed as suggestion for an objective scientific theory of human perception,the human capacity for deciphering phenomena,i.e.hermeneutics in its broadest sense.His added notes from the 1953 edition,all of which are comments,not corrections,imply that he never abandoned these thoughts on metaphysics,despite all of his utterances about a need to overcome it,and their popular interpretations to that effect.In his presentation,he further develops the colorful and intuitive style,an hermeneutic language,that he had created in his earlier work Being and Time.The logical functions of Dasein’s anatomical brain are performed by the logos machine,formerly the human soul,using top-down processing based on a global context,the noumenal cosmos which humans maintain internally.Heidegger’s 1942/43 winter semester lectures Parmenides extend in unbroken fashion his 1935 work,proving that he never abandoned,as is widely claimed,his metaphysical avenue of thought.
