Filters








40 Hits in 5.8 sec

Knowledge Sources for Constituent Parsing of German, a Morphologically Rich and Less-Configurational Language

Alexander Fraser, Helmut Schmid, Richárd Farkas, Renjing Wang, Hinrich Schütze
2013 Computational Linguistics  
One key question for MR&LC parsing is which type of parsing formalism to adopt, constituency or dependency.  ...  Then we examine both monolingual and bilingual approaches to parse reranking. Our reranking parser is the new state of the art in constituency parsing of the Tiger treebank.  ...  Other methods remove nodes and edges from the packed parse forest whose posterior probability is under a pre-defined threshold (Charniak and Johnson 2005) .  ... 
doi:10.1162/coli_a_00135 fatcat:c64swa66obhdjjpc7mleruqgfe

Phrase Dependency Machine Translation with Quasi-Synchronous Tree-to-Tree Features

Kevin Gimpel, Noah A. Smith
2014 Computational Linguistics  
We also present encouraging preliminary results on the use of unsupervised dependency parsing for syntax-based machine translation.  ...  We report statistically significant improvements over a phrasebased baseline on five of seven test sets across four language pairs.  ...  Also, by comparing the third baseline ("Moses, SSVM reranking") to our model, we are able to cleanly measure the contribution of our QPD features.  ... 
doi:10.1162/coli_a_00175 fatcat:e3hfi5iovvdrrhqbi76h3yrvwq

Feature Forest Models for Probabilistic HPSG Parsing

Yusuke Miyao, Jun'ichi Tsujii
2008 Computational Linguistics  
Feature forests are generic data structures that represent ambiguous trees in a packed forest structure. Feature forest models are maximum entropy models defined over feature forests.  ...  These methods have relied on the structure of the target problem, namely lattices or trees, and cannot be applied to graph structures including typed feature structures.  ...  The second and the third features are for the other two relations.  ... 
doi:10.1162/coli.2008.34.1.35 fatcat:5prj5nnnijcobiz5z337luuu6q

Message from the general chair

Benjamin C. Lee
2015 2015 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS)  
Our system gives a better performance than all the learning-based systems from the CoNLL-2011 shared task on the same dataset.  ...  Our end system outperforms the state-of-the-art baseline by 2 B 3 F1 points on non-transcript portion of the ACE 2004 dataset.  ...  We evaluated our reranking approach on German and English phrase structure parsing tasks and compared it to various state-of-the-art reranking approaches such as the perceptron-based forest reranker.  ... 
doi:10.1109/ispass.2015.7095776 dblp:conf/ispass/Lee15 fatcat:ehbed6nl6barfgs6pzwcvwxria

Getting Past the Language Gap: Innovations in Machine Translation [chapter]

Rodolfo Delmonte
2012 Mobile Speech and Advanced Natural Language Solutions  
Reranking is done on the basis of MERT, however this method is unstable.  ...  The system will look for alignments in adjacency of an already aligned word in case of misalignments: some words may come before or after another word depending on language grammar.  ...  During decoding, a packed forest of the source sentence is used as input, and the production rule tree fragments are applied to the packed forest.  ... 
doi:10.1007/978-1-4614-6018-3_6 fatcat:2njkc6meabhaxosl4wircumfjm

Survey of the State of the Art in Natural Language Generation: Core tasks, applications and evaluation [article]

Albert Gatt, Emiel Krahmer
2018 arXiv   pre-print
This survey therefore aims to (a) give an up-to-date synthesis of research on the core tasks in NLG and the architectures adopted in which such tasks are organised; (b) highlight a number of relatively  ...  other areas of artificial intelligence; (c) draw attention to the challenges in NLG evaluation, relating them to similar challenges faced in other areas of Natural Language Processing, with an emphasis on  ...  Whether or not nlg systems aim for variation in their output or not depends on the domain.  ... 
arXiv:1703.09902v4 fatcat:owx2fgo2bjej3b27ve2f3ledoe

Optimization for Statistical Machine Translation: A Survey

Graham Neubig, Taro Watanabe
2016 Computational Linguistics  
We also cover recent topics, including large-scale optimization, nonlinear models, domain-dependent optimization, and the effect of MT evaluation measures or search on optimization.  ...  In this article, we survey 12 years of research on optimization for SMT, from the seminal work on discriminative models (Och and Ney 2002) and minimum error rate training , to the most recent advances.  ...  It should be noted that different derivations in a k-best list may share the same translation due to the variation of phrases or rules in constructing a translation, e.g., the choice of "support the chair  ... 
doi:10.1162/coli_a_00241 fatcat:l6rxqwduozhzrmjia3frigppxu

Survey of the State of the Art in Natural Language Generation: Core tasks, applications and evaluation

Albert Gatt, Emiel Krahmer
2018 The Journal of Artificial Intelligence Research  
This survey therefore aims to (a) give an up-to-date synthesis of research on the core tasks in NLG and the architectures adopted in which such tasks are organised; (b) highlight a number of recent research  ...  synergies between NLG and other areas of artificial intelligence; (c) draw attention to the challenges in NLG evaluation, relating them to similar challenges faced in other areas of NLP, with an emphasis on  ...  Whether or not nlg systems aim for variation in their output or not depends on the domain.  ... 
doi:10.1613/jair.5477 fatcat:ycuteghjzncn7nx6pzkhzd6mn4

Language and other complex behaviors: Unifying characteristics, computational models, neural mechanisms

Shimon Edelman
2017 Language Sciences  
An analysis of the functional characteristics shared by complex sequential behaviors suggests that they all present a common overarching computational problem: dynamically controlled constrained navigation  ...  if the problem-level hypotheses are mistaken, without being recognized as such -as seems to be the case both in Marr's original field, vision, which is still widely and erroneously believed to hinge on  ...  a variation set.  ... 
doi:10.1016/j.langsci.2017.04.003 fatcat:kisve5cuqfb2nl42wioegqwvl4

Statistical Machine Translation by Generalized Parsing [article]

I. Dan Melamed, Wei Wang
2005 arXiv   pre-print
This article aims to reduce the conceptual complexity of such systems, in order to make them easier to design, implement, debug, use, study, understand, explain, modify, and improve.  ...  All these derivation trees, along with their probabilities, can be efficiently represented as a packed parse forest, rooted at the goal item.  ...  Assuming a fixed maximum fan-out f for the given grammar, the number of different spans in each inference depends on how many boundaries are shared between the antecedent items.  ... 
arXiv:cs/0407005v3 fatcat:yvp6qtiyibgqxaxbrmzbmueqm4

Efficient Query Processing for Scalable Web Search

Nicola Tonellotto, Craig Macdonald, Iadh Ounis
2018 Foundations and Trends in Information Retrieval  
Hence, there is a need for the development of efficient query processing infrastructures that make appropriate sacrifices in effectiveness in order to make gains in efficiency.  ...  Instead, this survey focuses on the general architecture of a search infrastructure as might be deployed within a single server.  ...  third.  ... 
doi:10.1561/1500000057 fatcat:wx53qhvfhnfwfc4hgdva5ypw3u

Labeling hierarchical phrase-based models without linguistic resources

Gideon Maillette de Buy Wenniger, Khalil Sima'an
2015 Machine Translation  
Long-range word order differences are a well-known problem for machine translation.  ...  In order to obtain alignment patterns that generalize well, we propose to decompose word alignments into trees over phrase pairs.  ...  Wilker Aziz for very helpful feedback on earlier versions of the manuscript. We also thank the anonymous reviewers for their helpful comments.  ... 
doi:10.1007/s10590-015-9177-0 fatcat:a3ncyottbvcdbfgzj75iujztzy

A scalable sparse Cholesky based approach for learning high-dimensional covariance matrices in ordered data

Kshitij Khare, Sang-Yun Oh, Syed Rahman, Bala Rajaratnam
2019 Machine Learning  
In such a situation, the expression levels of pairs of genes are dependent. However, gene X and gene Z do not share a direct relation, as they share only a common factor, gene Y.  ...  It should be noted that there is no perfect representation of a Bayesian network, and the JPD can be represented in different forms depending on the order of each node.  ...  By using learning based on a Bayesian network based on statistical dependencies, one can answer a wide range of queries, such as whether there is dependence between expression levels of a gene and the  ... 
doi:10.1007/s10994-019-05810-5 fatcat:nulmjvxvwjgojfoe2ywv3pjrpu

Realisation ranking and word ordering in the context of lesser resourced languages

Yasaman Motazedi
2022
thesis we are focusing on two subtasks of NLG: realisation ranking and word ordering .  ...  This involves making many decisions on the surface representation of sentences, including lexical selection, use of referring expressions, and word order relying on manually constructedgrammars.In this  ...  We propose a future study on incorporating the grammaticality constraints into Recurrent Neural Network (RNN)s to improve the produced text.  ... 
doi:10.25949/19435181 fatcat:zl6rxqzgkrh4jo3it5hzmbnhsq

Identification of Ebola Virus Inhibitors Targeting GP2 using Principles of Molecular Mimicry

Courtney D. Singleton, Monica S. Humby, Hyun Ah Yi, Robert C. Rizzo, Amy Jacobs
2019 Journal of Virology  
Based on assessment of inhibitory activity, cytotoxicity, and target specificity, four promising candidates emerged with IC50 values in the 3-26 μM range.  ...  FIG 8 8 Infectivity of candidate compounds on VSV-G-pseudotyped virus entry ordered by infectivity.  ...  Third, over 250 ps, the system was heated from 50 to 300K.  ... 
doi:10.1128/jvi.00676-19 pmid:31092576 pmcid:PMC6639268 fatcat:t4vxchb5nzhh5pt5bnugge7itu
« Previous Showing results 1 — 15 out of 40 results