multi source translation methods
play

Multi-Source Translation Methods Lane Schwartz lane@cs.umn.edu - PowerPoint PPT Presentation

Sep 10, 2022 •219 likes •838 views

Related Work Oracle Experiment Revisiting Och and Ney (2001) Multi-Source Translation Methods Lane Schwartz lane@cs.umn.edu University of Minnesota 23 Oct 2008 Lane Schwartz lane@cs.umn.edu Multi-Source Translation Methods Related Work

  1. Related Work Oracle Experiment Revisiting Och and Ney (2001) Multi-Source Translation Methods Lane Schwartz lane@cs.umn.edu University of Minnesota 23 Oct 2008 Lane Schwartz lane@cs.umn.edu Multi-Source Translation Methods

  2. Related Work Oracle Experiment Revisiting Och and Ney (2001) Related Work Oracle Experiment Revisiting Och and Ney (2001) Lane Schwartz lane@cs.umn.edu Multi-Source Translation Methods

  3. Related Work Oracle Experiment Revisiting Och and Ney (2001) Motivation Principle of Translational Promiscuity: If a document is translated into more than 1 language, it will likely be translated into many more languages. ◮ Translate into first n target languages by hand ◮ Translate into remaining target languages using MT Lane Schwartz lane@cs.umn.edu Multi-Source Translation Methods

  4. Related Work Oracle Experiment Revisiting Och and Ney (2001) Motivation Principle of Translational Promiscuity: If a document is translated into more than 1 language, it will likely be translated into many more languages. ◮ Translate into first n target languages by hand ◮ Translate into remaining target languages using MT Lane Schwartz lane@cs.umn.edu Multi-Source Translation Methods

  5. Related Work Oracle Experiment Revisiting Och and Ney (2001) Motivation Principle of Translational Promiscuity: If a document is translated into more than 1 language, it will likely be translated into many more languages. ◮ Translate into first n target languages by hand ◮ Translate into remaining target languages using MT Lane Schwartz lane@cs.umn.edu Multi-Source Translation Methods

  6. Related Work Oracle Experiment Revisiting Och and Ney (2001) Related Work Can using multiple sources of information improve translation? ◮ Lattice inputs ◮ Consensus decoding ◮ Hypothesis ranking Lane Schwartz lane@cs.umn.edu Multi-Source Translation Methods

  7. Related Work Oracle Experiment Revisiting Och and Ney (2001) Related Work Can using multiple sources of information improve translation? ◮ Lattice inputs ◮ Consensus decoding ◮ Hypothesis ranking Lane Schwartz lane@cs.umn.edu Multi-Source Translation Methods

  8. Related Work Oracle Experiment Revisiting Och and Ney (2001) Related Work Can using multiple sources of information improve translation? ◮ Lattice inputs ◮ Consensus decoding ◮ Hypothesis ranking Lane Schwartz lane@cs.umn.edu Multi-Source Translation Methods

  9. Related Work Oracle Experiment Revisiting Och and Ney (2001) Related Work Can using multiple sources of information improve translation? ◮ Lattice inputs ◮ Consensus decoding ◮ Hypothesis ranking Lane Schwartz lane@cs.umn.edu Multi-Source Translation Methods

  10. Related Work Oracle Experiment Revisiting Och and Ney (2001) Related Work Can using multiple sources of information improve translation? ◮ Lattice inputs ◮ Consensus decoding ◮ Hypothesis ranking Lane Schwartz lane@cs.umn.edu Multi-Source Translation Methods

  11. Related Work Oracle Experiment Revisiting Och and Ney (2001) Related Work — Lattice Input ◮ Begin with alternate representations of a source sentence Chinese word segmentations Arabic morphological analyses ◮ Align alternate representations into a word lattice ◮ Use standard decoding algorithms, modified to accept lattice input (Dyer et al., 2008) ◮ Can be extended to accept multilingual inputs (ongoing work by Josh Schroeder) Lane Schwartz lane@cs.umn.edu Multi-Source Translation Methods

  12. Related Work Oracle Experiment Revisiting Och and Ney (2001) Related Work — Lattice Input ◮ Begin with alternate representations of a source sentence Chinese word segmentations Arabic morphological analyses ◮ Align alternate representations into a word lattice ◮ Use standard decoding algorithms, modified to accept lattice input (Dyer et al., 2008) ◮ Can be extended to accept multilingual inputs (ongoing work by Josh Schroeder) Lane Schwartz lane@cs.umn.edu Multi-Source Translation Methods

  13. Related Work Oracle Experiment Revisiting Och and Ney (2001) Related Work — Lattice Input ◮ Begin with alternate representations of a source sentence Chinese word segmentations Arabic morphological analyses ◮ Align alternate representations into a word lattice ◮ Use standard decoding algorithms, modified to accept lattice input (Dyer et al., 2008) ◮ Can be extended to accept multilingual inputs (ongoing work by Josh Schroeder) Lane Schwartz lane@cs.umn.edu Multi-Source Translation Methods

  14. Related Work Oracle Experiment Revisiting Och and Ney (2001) Related Work — Lattice Input ◮ Begin with alternate representations of a source sentence Chinese word segmentations Arabic morphological analyses ◮ Align alternate representations into a word lattice ◮ Use standard decoding algorithms, modified to accept lattice input (Dyer et al., 2008) ◮ Can be extended to accept multilingual inputs (ongoing work by Josh Schroeder) Lane Schwartz lane@cs.umn.edu Multi-Source Translation Methods

  15. Related Work Oracle Experiment Revisiting Och and Ney (2001) Related Work — Lattice Input ◮ Begin with alternate representations of a source sentence Chinese word segmentations Arabic morphological analyses ◮ Align alternate representations into a word lattice ◮ Use standard decoding algorithms, modified to accept lattice input (Dyer et al., 2008) ◮ Can be extended to accept multilingual inputs (ongoing work by Josh Schroeder) Lane Schwartz lane@cs.umn.edu Multi-Source Translation Methods

  16. Related Work Oracle Experiment Revisiting Och and Ney (2001) Related Work — Consensus Decoding ◮ Given a set of translations, find a consensus translation ◮ Bilingual consensus decoding (Frederking and Nirenburg, 1994; Bangalore et al., 2001; Jayaraman and Lavie, 2005; Rosti et al., 2007) ◮ Translate source text using n different systems ◮ Align the n output hypotheses into a weighted word lattice ◮ Intersect word lattice with n-gram language model ◮ Multilingual consensus decoding ◮ Matusov et al. (2006) ◮ Japanese and Chinese into English ◮ 4.8 BLEU improvement Lane Schwartz lane@cs.umn.edu Multi-Source Translation Methods

  17. Related Work Oracle Experiment Revisiting Och and Ney (2001) Related Work — Consensus Decoding ◮ Given a set of translations, find a consensus translation ◮ Bilingual consensus decoding (Frederking and Nirenburg, 1994; Bangalore et al., 2001; Jayaraman and Lavie, 2005; Rosti et al., 2007) ◮ Translate source text using n different systems ◮ Align the n output hypotheses into a weighted word lattice ◮ Intersect word lattice with n-gram language model ◮ Multilingual consensus decoding ◮ Matusov et al. (2006) ◮ Japanese and Chinese into English ◮ 4.8 BLEU improvement Lane Schwartz lane@cs.umn.edu Multi-Source Translation Methods

  18. Related Work Oracle Experiment Revisiting Och and Ney (2001) Related Work — Consensus Decoding ◮ Given a set of translations, find a consensus translation ◮ Bilingual consensus decoding (Frederking and Nirenburg, 1994; Bangalore et al., 2001; Jayaraman and Lavie, 2005; Rosti et al., 2007) ◮ Translate source text using n different systems ◮ Align the n output hypotheses into a weighted word lattice ◮ Intersect word lattice with n-gram language model ◮ Multilingual consensus decoding ◮ Matusov et al. (2006) ◮ Japanese and Chinese into English ◮ 4.8 BLEU improvement Lane Schwartz lane@cs.umn.edu Multi-Source Translation Methods

  19. Related Work Oracle Experiment Revisiting Och and Ney (2001) Related Work — Consensus Decoding ◮ Given a set of translations, find a consensus translation ◮ Bilingual consensus decoding (Frederking and Nirenburg, 1994; Bangalore et al., 2001; Jayaraman and Lavie, 2005; Rosti et al., 2007) ◮ Translate source text using n different systems ◮ Align the n output hypotheses into a weighted word lattice ◮ Intersect word lattice with n-gram language model ◮ Multilingual consensus decoding ◮ Matusov et al. (2006) ◮ Japanese and Chinese into English ◮ 4.8 BLEU improvement Lane Schwartz lane@cs.umn.edu Multi-Source Translation Methods

  20. Related Work Oracle Experiment Revisiting Och and Ney (2001) Related Work — Consensus Decoding ◮ Given a set of translations, find a consensus translation ◮ Bilingual consensus decoding (Frederking and Nirenburg, 1994; Bangalore et al., 2001; Jayaraman and Lavie, 2005; Rosti et al., 2007) ◮ Translate source text using n different systems ◮ Align the n output hypotheses into a weighted word lattice ◮ Intersect word lattice with n-gram language model ◮ Multilingual consensus decoding ◮ Matusov et al. (2006) ◮ Japanese and Chinese into English ◮ 4.8 BLEU improvement Lane Schwartz lane@cs.umn.edu Multi-Source Translation Methods

  21. Related Work Oracle Experiment Revisiting Och and Ney (2001) Related Work — Consensus Decoding ◮ Given a set of translations, find a consensus translation ◮ Bilingual consensus decoding (Frederking and Nirenburg, 1994; Bangalore et al., 2001; Jayaraman and Lavie, 2005; Rosti et al., 2007) ◮ Translate source text using n different systems ◮ Align the n output hypotheses into a weighted word lattice ◮ Intersect word lattice with n-gram language model ◮ Multilingual consensus decoding ◮ Matusov et al. (2006) ◮ Japanese and Chinese into English ◮ 4.8 BLEU improvement Lane Schwartz lane@cs.umn.edu Multi-Source Translation Methods

Recommend

Adaptive Multi-pass Decoder for Neural Machine Translation EMNLP 2018

Adaptive Multi-pass Decoder for Neural Machine Translation EMNLP 2018

Adaptive Multi-pass Decoder for Neural Machine Translation EMNLP 2018 http://aclweb.org/anthology/D18-1048 Neural Machine Translation(NMT) The encoder-decoder are widely used in neural machine translation the encoder transforms the source

121 views • 11 slides
Inferring Translation Candidates for Multilingual Dictionary Generation with Multi-Way Neural

Inferring Translation Candidates for Multilingual Dictionary Generation with Multi-Way Neural

Inferring Translation Candidates for Multilingual Dictionary Generation with Multi-Way Neural Machine Translation Mihael Arcan, Daniel Torregrosa*, Sina Ahmadi* and John P. McCrae This publication has emanated from research supported in part

591 views • 42 slides
Atomicity via Source-to-Source Translation Benjamin Hindman Dan Grossman University of

Atomicity via Source-to-Source Translation Benjamin Hindman Dan Grossman University of

Atomicity via Source-to-Source Translation Benjamin Hindman Dan Grossman University of Washington 22 October 2006 Atomic An easier-to-use and harder-to-implement primitive void deposit(int x){ void deposit(int x){ synchronized(this){

516 views • 33 slides
SE350: Operating Systems Lecture 10: Address Translation Outline Multi-step processing of

SE350: Operating Systems Lecture 10: Address Translation Outline Multi-step processing of

SE350: Operating Systems Lecture 10: Address Translation Outline Multi-step processing of programs Virtual to physical address translation Segment mapping Page tables Multi-level tables Inverted page table Virtualizing

593 views • 56 slides
4. Source Encoding Methods Called also entropy coders , because the methods try to get

4. Source Encoding Methods Called also entropy coders , because the methods try to get

4. Source Encoding Methods Called also entropy coders , because the methods try to get close to the entropy (i.e. lower bound of compression). statistical coders , because the methods assume the probability distribution of the source

1.06k views • 33 slides
and Retrieval Source: H. Jegou Source: H. Jegou Source: H. Jegou Source: H. Jegou Source: H.

and Retrieval Source: H. Jegou Source: H. Jegou Source: H. Jegou Source: H. Jegou Source: H.

Semantic Image Indexing and Retrieval Source: H. Jegou Source: H. Jegou Source: H. Jegou Source: H. Jegou Source: H. Jegou Source: H. Jegou Source: H. Jegou Source: H. Jegou Outline State of the nation Early description methods

2.13k views • 130 slides
SDS Aplications - Speech-to-speech translation - Anca Burducea May 28, 2015 S2S Translation

SDS Aplications - Speech-to-speech translation - Anca Burducea May 28, 2015 S2S Translation

SDS Aplications - Speech-to-speech translation - Anca Burducea May 28, 2015 S2S Translation Three independent tasks: S s T s T t S t S s = speech source T s = text source T t = text target S t = speech target S2S Translation S s

400 views • 23 slides
Models and methods for plan diagnosis Brian Sulcer 10 March 2009 Source Roos, N &

Models and methods for plan diagnosis Brian Sulcer 10 March 2009 Source Roos, N &

Models and methods for plan diagnosis Brian Sulcer 10 March 2009 Source Roos, N & Witteveen, C. (2007). Models and methods for plan diagnosis. Autonomous Agents and Multi-Agent Systems 19(1):30-52 Model-based plan diagnosis plan

512 views • 16 slides
THE TRANSLATION TOOL EXPERTS Try it, love it GETTING STARTED Dj Vu X3 WITH TEAM SERVER

THE TRANSLATION TOOL EXPERTS Try it, love it GETTING STARTED Dj Vu X3 WITH TEAM SERVER

THE TRANSLATION TOOL EXPERTS Try it, love it GETTING STARTED Dj Vu X3 WITH TEAM SERVER Dj Vu TEAMserver , MAKES IT POSSIBLE FOR YOUR TRANSLATION TEAM , WORKING ON EXTENSIVE, MULTILINGUAL AND MULTI-SITE TRANSLATION PROJECTS, TO SHARE

349 views • 21 slides
Translation Model Parallel corpus source target translation e f phrase phrase features

Translation Model Parallel corpus source target translation e f phrase phrase features

Translation Model Parallel corpus source target translation e f phrase phrase features Reranking Model Monolingual feature corpus Language Model weights f Held-out e f parallel corpus

1.1k views • 63 slides
Multi - Grid Methods Press et al., Numerical Recipes in C , Section 19.6

Multi - Grid Methods Press et al., Numerical Recipes in C , Section 19.6

Multi - Grid Methods Press et al., Numerical Recipes in C , Section 19.6 http://www.nrbook.com/a/bookcpdf.php A Multigrid Tutorial, http://www.llnl.gov/casc/people/henson/mgtut/welcome.html Tuesday, June 26, 2007 1 Multi - Grid Methods

407 views • 22 slides
Open-source PortugueseSpanish machine translation C. Armentano-Oller 1 , R.C. Carrasco 1 , 2 ,

Open-source PortugueseSpanish machine translation C. Armentano-Oller 1 , R.C. Carrasco 1 , 2 ,

The Apertium MT toolbox Data for the pt es pair Concluding remarks Open-source PortugueseSpanish machine translation C. Armentano-Oller 1 , R.C. Carrasco 1 , 2 , A.M. Corb-Bellot 1 , M.L. Forcada 1 , 2 , M. Ginest-Rosell 1 , S.

338 views • 30 slides
Multi-Source Adjustment of Multi-Layer Annotation: the Bits of Wisdom Approach Kilian Evang 20

Multi-Source Adjustment of Multi-Layer Annotation: the Bits of Wisdom Approach Kilian Evang 20

Multi-Source Adjustment of Multi-Layer Annotation: the Bits of Wisdom Approach Multi-Source Adjustment of Multi-Layer Annotation: the Bits of Wisdom Approach Kilian Evang 20 January 2012 http://gmb.let.rug.nl 1/1 Multi-Source Adjustment of

611 views • 14 slides
Empirical Methods in Natural Language Processing Lecture 19 Machine translation (VI): Factored

Empirical Methods in Natural Language Processing Lecture 19 Machine translation (VI): Factored

Empirical Methods in Natural Language Processing Lecture 19 Machine translation (VI): Factored Translation Models Philipp Koehn 10 March 2008 Philipp Koehn EMNLP Lecture 19 10 March 2008 1 Statistical machine translation today Best

344 views • 10 slides
Sequence to Sequence Models for Machine Translation CMSC 723 / LING 723 / INST 725 Marine

Sequence to Sequence Models for Machine Translation CMSC 723 / LING 723 / INST 725 Marine

Sequence to Sequence Models for Machine Translation CMSC 723 / LING 723 / INST 725 Marine Carpuat Slides & figure credits: Graham Neubig Machine Translation 3 problems Translation system Input: source sentence F Output:

349 views • 20 slides
Machine Translation using Deep Learning Methods Max Fomin Michael Zolotov Sequence to Sequence

Machine Translation using Deep Learning Methods Max Fomin Michael Zolotov Sequence to Sequence

Machine Translation using Deep Learning Methods Max Fomin Michael Zolotov Sequence to Sequence Learning with Neural Networks Learning Phrase Representations using RNN Encoder Decoder for Statistical Machine Translation Topics

909 views • 41 slides
GlobalSight is an open-source translation management platform that manages the end-to-end

GlobalSight is an open-source translation management platform that manages the end-to-end

GlobalSight is an open-source translation management platform that manages the end-to-end localization process . GlobalSight provides a robust, easy-to- access platform for managing, translating and delivering global content.

72 views • 6 slides
FOR GEOMETRIC MULTI-GRID METHODS Nikolay Sakharnykh (NVIDIA), Dusan Stosic (UFPE), 4/4/2016

FOR GEOMETRIC MULTI-GRID METHODS Nikolay Sakharnykh (NVIDIA), Dusan Stosic (UFPE), 4/4/2016

April 4-7, 2016 | Silicon Valley HIGH-ORDER DISCRETIZATIONS FOR GEOMETRIC MULTI-GRID METHODS Nikolay Sakharnykh (NVIDIA), Dusan Stosic (UFPE), 4/4/2016 MULTI-GRID METHODS Introduction Multi-grid solves elliptic PDEs (Ax=b) using a

684 views • 44 slides
A Simplistic Translation Scheme CS429: Computer Organization and Architecture m.c ASCII source

A Simplistic Translation Scheme CS429: Computer Organization and Architecture m.c ASCII source

A Simplistic Translation Scheme CS429: Computer Organization and Architecture m.c ASCII source file Problems: Linking I & II Efficiency: small change requires complete Compiler re-compilation. Dr. Bill Young Modularity: hard to share

411 views • 5 slides
A Widely Used Machine Translation Service and its Migration to a Free/Open-Source Solution: the

A Widely Used Machine Translation Service and its Migration to a Free/Open-Source Solution: the

A Widely Used Machine Translation Service and its Migration to a Free/Open-Source Solution: the Case of Softcatal Xavier Ivars-Ribes Victor M. Snchez-Cartagena II FreeRBMT (Barcelona) January 21, 2011 Table of Contents Brief History of

417 views • 23 slides
1 Translation model Language model Dictionaries used Languages Name #Entries Type P(S|T)

1 Translation model Language model Dictionaries used Languages Name #Entries Type P(S|T)

Motivation Classification of CLIR methods Cross-Language IR at We developed an automatic transliteration query translation method University of Tsukuba method for Japanese and English CLIR document translation method Automatic

210 views • 3 slides
Multi-column Substring Matching For Database Schema Translation Robert H. Warren 1 Dr. Frank Wm

Multi-column Substring Matching For Database Schema Translation Robert H. Warren 1 Dr. Frank Wm

Research Motivation Proposed Approach Experimental Results Algorithmic Analysis Conclusion Multi-column Substring Matching For Database Schema Translation Robert H. Warren 1 Dr. Frank Wm Tompa 1 1 David R. Cheriton School of Computer Science

520 views • 38 slides
AN OPEN-SOURCE FRAMEWORK FOR INTEGRATING MULTI-SOURCE LAYOUT AND TEXT RECOGNITION TOOLS INTO

AN OPEN-SOURCE FRAMEWORK FOR INTEGRATING MULTI-SOURCE LAYOUT AND TEXT RECOGNITION TOOLS INTO

AN OPEN-SOURCE FRAMEWORK FOR INTEGRATING MULTI-SOURCE LAYOUT AND TEXT RECOGNITION TOOLS INTO SCALABLE OCR WORKFLOWS KAY-MICHAEL WRZNER KONSTANTIN BAIERER 1 . 1 Bibliotheca Baltica 2018 Rostock 2018-10-05 OVERVIEW 1. Why OCR-D 2. The

776 views • 48 slides
Multi-wavelength cross-correlation methods Mara Salvato (With a thanks to F. Guglielmetti &

Multi-wavelength cross-correlation methods Mara Salvato (With a thanks to F. Guglielmetti &

Multi-wavelength cross-correlation methods Mara Salvato (With a thanks to F. Guglielmetti & M. Brusa) 1 Structure of the talk: Why multi wavelength cross-matches are important All what can go wrong Methods descriptions

424 views • 21 slides

More recommend