https://wiki.ufal.ms.mff.cuni.cz/ 2026-04-19T01:32:53+00:00 https://wiki.ufal.ms.mff.cuni.cz/ https://wiki.ufal.ms.mff.cuni.cz/lib/tpl/ufal/images/favicon.ico text/html 2012-11-19T18:20:02+00:00 Anonymous (anonymous@undisclosed.example.com) https://wiki.ufal.ms.mff.cuni.cz/courses:rg:2012:alignment-by-agreement?rev=1353345602&do=diff Alignment by Agreement Percy Liang, Ben Taskar, Dan Klein, link Section 2 -- discussion about previous alignment models IBM Models 1, 2 and HMM alignment model * all decompose into a product of p_d (distortion probability) and p_t (translation probability) text/html 2012-05-14T22:10:52+00:00 Anonymous (anonymous@undisclosed.example.com) https://wiki.ufal.ms.mff.cuni.cz/courses:rg:2012:applying-morphology-to-mt?rev=1337026252&do=diff Applying Morphology Generation Models to Machine Translation paper by: Kristina Toutanova, Hisami Suzuki, and Achim Ruopp presentend by: Amir Kamran report by: Martin Popel Comments * Two base MT systems (treelet and phrasal) were improved by applying models that generate word forms from target-language stems and source-language sentence. These models are MEMM trained independently on the base MT. text/html 2012-06-03T21:01:01+00:00 Anonymous (anonymous@undisclosed.example.com) https://wiki.ufal.ms.mff.cuni.cz/courses:rg:2012:atreport?rev=1338750061&do=diff Semantic Taxonomy Induction from Heterogenous Evidence Introduction - related methods (WordNet -- hand-made, CYC) - hand-made patterns “filled in” by words that satisfy them (automaticaly) - “such NP(y) as NP (x)” => y is hypernym of x (reversed in the paper! probably a copy-paste error) - most methods disregard ambiguity (rose bush) text/html 2012-12-16T23:44:57+00:00 Anonymous (anonymous@undisclosed.example.com) https://wiki.ufal.ms.mff.cuni.cz/courses:rg:2012:distributed-perceptron?rev=1355697897&do=diff Distributed Training Strategies for the Structured Perceptron - RG report - UNDER CONSTRUCTION Presentation 3 Structured Perceptron * In unstructured perceptron, you are trying to separate two sets of with hyperplane. See Question 1 for the algorithm. In training phase, you iterate your training data and adjust the hyperplane every time you make a mistake. text/html 2012-10-29T20:30:57+00:00 Anonymous (anonymous@undisclosed.example.com) https://wiki.ufal.ms.mff.cuni.cz/courses:rg:2012:encouraging-consistent-translation-bushra?rev=1351539057&do=diff Introduction: This paper emphasizes on using “one translation per discourse” heuristic in hierarchical phrase-based machine translation after getting motivated by “one sense per discourse” heuristic in Word Sense Disambiguation. A document (domain specific) is treated as a discourse unit in this paradigm. A novel approach of forced decoding is used to implement the heuristic in three different ways in machine translation system. Experiments are performed on Arabic-English and Chinese-English la… text/html 2012-10-23T11:04:53+00:00 Anonymous (anonymous@undisclosed.example.com) https://wiki.ufal.ms.mff.cuni.cz/courses:rg:2012:encouraging-consistent-translation?rev=1350983093&do=diff Encouraging Consistent Translation Choices Ferhan Ture, Douglas W. Oard, and Philip Resnik NAACL 2012 PDF Outline -- discussion The list of discussed topics follows the outline of the paper: Sec. 2. Related Work Differences from Carpuat 2009 * It is different: the decoder just gets additional features, but the decision is up to it text/html 2012-03-26T18:51:16+00:00 Anonymous (anonymous@undisclosed.example.com) https://wiki.ufal.ms.mff.cuni.cz/courses:rg:2012:jodaiberreport?rev=1332780676&do=diff text/html 2012-03-12T22:59:35+00:00 Anonymous (anonymous@undisclosed.example.com) https://wiki.ufal.ms.mff.cuni.cz/courses:rg:2012:longdtreport?rev=1331589575&do=diff Faster and Smaller N-Gram Language Model Presenter : Joachim Daiber Reporter: Long DT Date : 12-March-2012 Overview The talk is mainly about techniques to improve performance of N-gram language model. How it will run faster and use smaller amount of memory. text/html 2012-11-13T16:25:16+00:00 Anonymous (anonymous@undisclosed.example.com) https://wiki.ufal.ms.mff.cuni.cz/courses:rg:2012:meant?rev=1352820316&do=diff MEANT: An inexpensive, high-accuracy, semi-automatic metric for evaluating translation utility via semantic frames Chi-kiu Lo and Dekai Wu ACL 2011 <http://www.aclweb.org/anthology/P11-1023> Presented by Petr Jankovský Report by Rudolf Rosa The paper was widely discussed throughout the whole session. The report tries to divide the points discussed in correspondence to the sections of the paper. text/html 2012-12-03T10:28:46+00:00 Anonymous (anonymous@undisclosed.example.com) https://wiki.ufal.ms.mff.cuni.cz/courses:rg:2012:riezler-iii?rev=1354526926&do=diff Martin's questions 1) How would you implement approximate randomization for BLEU based on Figure 1, namely the part "Shuffle variable tuples between system X and Y with probability 0.5"? What are the variable tuples? Can you write a more detailed pseudo (or C,Java,Perl,...) code? How would you implement the next part "Compute pseudo-statistic |S_Xr − S_Yr | on shuffled data"? text/html 2012-09-17T01:38:37+00:00 Anonymous (anonymous@undisclosed.example.com) https://wiki.ufal.ms.mff.cuni.cz/courses:rg:2012:rosareport?rev=1347838717&do=diff Training Phrase Translation Models with Leaving-One-Out paper by Joern Wuebker, Arne Mauser and Hermann Ney presented by Bushra Jawaid report by Rudolf Rosa Presentation The paper was well presented. Bushra talked about the paper in great detail, even including some information from the related papers. However, this lead to a time shortage towards the end of the presentation. text/html 2012-09-25T14:48:58+00:00 Anonymous (anonymous@undisclosed.example.com) https://wiki.ufal.ms.mff.cuni.cz/courses:rg:2012:searn-in-practice?rev=1348577338&do=diff Searn in Practice paper by: Hal Daumé III, John Langford and Daniel Marcu presented by: Martin Popel report by: Petra Galuščáková Comments * Searn (stands for search-learn) is a novel algorithm for solving hard structured prediction problems. A structured prediction problem D is a cost-sensitive classification problem where Y has structure: elements y ∈ Y decompose into variable-length vectors (y text/html 2013-01-03T22:38:32+00:00 Anonymous (anonymous@undisclosed.example.com) https://wiki.ufal.ms.mff.cuni.cz/courses:rg:2012:segments?rev=1357249112&do=diff Introduction, Motivation, Segments We introduced the basic idea of Czech sentence segmentation and the Czech sentence boundaries. We showed the segmentation chart on an example. Experiments with Automatic Identification of Segmentation Charts How to Obtain Segments from Syntactic Tree? text/html 2013-12-02T22:18:19+00:00 Anonymous (anonymous@undisclosed.example.com) https://wiki.ufal.ms.mff.cuni.cz/courses:rg:2012:sigtest-mt-zilka?rev=1386019099&do=diff Questions Question 1 REF: John thinks he loves Mary MT1: John thinks he loves Mary MT2: John knows he loves Mary MT3: John thinks he loves RG Given a test corpus with this one sentence, what are the BLEU scores of the three systems based on formulas (1) and (2)? text/html 2012-11-12T17:43:22+00:00 Anonymous (anonymous@undisclosed.example.com) https://wiki.ufal.ms.mff.cuni.cz/courses:rg:2012:sigtest-mt?rev=1352738602&do=diff Statistical Significance Tests for Machine Translation Evaluation Koehn, EMNLP 2004, link Questions 1) BLEU_MT1 = 1, BLEU_MT2 = 0 (or undefined) BLEU_MT3 = 0.2 (according to the formula in the paper, incorrect) It should be exp(1/4(ln(4/5) + ln(3/4) + ln(2/3) + ln(1/2))) = 0.668 text/html 2012-10-29T17:50:57+00:00 Anonymous (anonymous@undisclosed.example.com) https://wiki.ufal.ms.mff.cuni.cz/courses:rg:2012:soft-synt-consts-for-hierarchiacl-phrase-based-trans?rev=1351529457&do=diff Soft Syntactic Constraints for Hierarchical Phrase-based Translation Using Latent Syntactic Distributions Zhongqiang Huang, Martin Čmejrek and Bowen Zhou Conference on Empirical Methods in NLP, 2010 PDF Presented by Jindřich Helcl Report by Petr Jankovský text/html 2012-10-12T14:17:42+00:00 Anonymous (anonymous@undisclosed.example.com) https://wiki.ufal.ms.mff.cuni.cz/courses:rg:2012:spe-for-smt?rev=1350044262&do=diff Statistical Post-Editing for a Statistical MT System Hanna Béchara, Yanjun Ma, Josef van Genabith MT Summit 2011 PDF Presented by Rudolf Rosa Report by Jindřich Helcl Introduction This article was about statistical post-editing on results of a statistical machine translation system. The most interesting part on this article was that authors claim that they achieved improvement of about 2 BLEU score points by pipelining two statistical MT systems, which was until then considered useless. text/html 2012-12-01T23:09:45+00:00 Anonymous (anonymous@undisclosed.example.com) https://wiki.ufal.ms.mff.cuni.cz/courses:rg:2012:stat-nlg?rev=1354399785&do=diff Phrase-based Statistical Language Generation using Graphical Models and Active Learning François Mairesse, Milica Gašić, Filip Jurčíček, Simon Keizer, Blaise Thomson, Kai Yu, Steve Young ACL 2010 <http://aclweb.org/anthology-new/P/P10/P10-1157.pdf> Presented by Ondřej Dušek Report by Honza Václ text/html 2012-05-10T14:49:03+00:00 Anonymous (anonymous@undisclosed.example.com) https://wiki.ufal.ms.mff.cuni.cz/courses:rg:2012:the-unreasonable-effectiveness-of-data-paper?rev=1336654143&do=diff The Unreasonable Effectiveness of Data * PDF * Peter Norvig - The Unreasonable Effectiveness of Data - Youtube Related Reading * Data-Intensive Text Processing with MapReduce - chapter 1