MT Lunch Seminar (LTI, CMU)

Training Machine Translation with a Second-Order Taylor Approximation of Weighted Translation Instances

2011-09-07T16:35:00.000-07:00

Title: Training Machine Translation with a Second-Order Taylor Approximation of Weighted Translation Instances
Speaker: Aaron Phillips
When: Tuesday, September 13, 12:00 Noon to 1:00pm.
Where: GHC 6501

Abstract: The Cunei Machine Translation Platform is an open-source MT system designed to model instances of translation. One of the challenges to this approach is effective training. We describe two techniques that improve the training procedure and allow us to leverage the strengths of instance-based modeling. First, during training we approximate our model with a second-order Taylor series. Second, we discount models based on the magnitude of their approximation. By reducing error in training, our model now consistently outperforms the standard SMT model with gains ranging from 0.51 to 3.77 BLEU on German-English and Czech-English test sets.

Syntax-to-Morphology Mapping in Factored Phrase-Based SMT (English and Turkish)

2011-05-16T13:19:00.001-07:00

Title: Syntax-to-Morphology Mapping in Factored Phrase-Based
Statistical Machine Translation between English and Turkish

Speaker: Reyyan Yeniterzi

When: Tuesday, May 17 at 12:15pm
Where: GHC 6501

Abstract:

Motivated by the observation that many local and some nonlocal
syntactic structures in English essentially map to morphologically
complex words in Turkish, a new approach which is called
syntax-to-morphology mapping was introduced recently (Yeniterzi and
Oflazer, 2010). This approach maps syntactic structures in English to
complex words in Turkish directly. It mainly recognizes certain local
and nonlocal syntactic structures on the English side and packages
those structures and attach to heads to obtain parallel morphological
structures.

With the help of this method, one can identify and reorganize phrases
on the English side, to align English syntax to Turkish morphology.
Furthermore with this method, continuous and discontinuous variants of
certain (syntactic) source phrases can be conflated during the SMT
phrase extraction process. Since most function words encoding syntax
are now abstracted into complex tags, the length of the English
sentences can be dramatically reduced.

The initial experiments were performed on English-to-Turkish SMT
system. In this project, we built upon this initial system by doing
lexical reordering and data augmentation. Furthermore we also applied
syntax-to-morphology mapping to a Turkish-to-English SMT system for
the first time.

This is joint work with Kemal Oflazer from Qatar CMU. It was presented
in the Machine Translation and Morphologically-rich Languages Research
Workshop at Haifa, Israel in January, 2011.

Better Hypothesis Testing for Statistical Machine Translation: Controlling for Optimizer Instability

2011-05-03T13:41:00.001-07:00

Title: Better Hypothesis Testing for Statistical Machine Translation:
Controlling for Optimizer Instability
Speaker: Jonathan Clark
When: Tuesday, 4/19 at Noon

Abstract:
In statistical machine translation, a researcher seeks to determine
whether some innovation (e.g., a new feature, model, or inference
algorithm) improves translation quality in comparison to a baseline
system. To answer this question, he runs an experiment to evaluate the
behavior of the two systems on held-out data. In this paper, we
consider how to make such experiments more statistically reliable. We
provide a systematic analysis of the effects of optimizer instability
(an extraneous variable that is seldom controlled for) on experimental
outcomes, and make recommendations for reporting results more
accurately.

This is joint work with Chris Dyer, Alon Lavie, and Noah Smith. It was
recently accepted for publication as an ACL short paper.

Qin Gao: Expanding parallel corpora for machine translation

2011-03-02T12:29:00.000-08:00

Speaker: Qin Gao
When: at noon, March 8, 2011
Where: GHC 4405

We present an approach of expanding parallel corpora for machine translation. By utilizing Semantic role labeling (SRL) on one side of the language pair, we extract SRL substitution rules from existing parallel corpus. The rules are then used for generating new sentence pairs. An SVM classifier is built to filter the generated sentence pairs. The filtered corpus is used for training phrase-based translation models, which can be used directly in translation tasks or combined with baseline models. Experiment results on Chinese-English machine translation tasks show an average improvement of 0.45 BLEU and 1.22 TER points across 5 different NIST test sets.

Machine Translation and Computer-Assisted Translation

2011-01-13T17:34:00.000-08:00

Title: Prospects for Integrating Machine Translation and Computer-Assisted Translation in the Translation Industry

Speaker: Gregory M. Shreve from the Department of Modern and Classical Language Studies at Kent State University and colleagues
Location: GHC 6115
Time: 12:30 pm, 14 Jan 2011

The speaker's CV can be found at http://www.kent.edu/mcls/faculty/mcls_shreve.cfm.

Efficient Language Model Inference - Kenneth Heafield

2010-12-13T12:21:00.000-08:00

Title: Efficient Language Model Inference
Who? Kenneth Heafield
When? Tuesday, December 21 @ Noon
Where? GHC 4405

In GHC 4405 at noon on Tuesday Dec 21, I will give a speaking
requirement talk on Efficient Language Model Inference. As this is also
a MT Lunch, there will be free lunch.

If you're using SRILM, come to my talk to reduce your memory consumption
by 86% while reducing CPU time by 16%. Users of IRSTLM should come for
the same reason; the code uses 42% less memory and 19% less CPU.

Language models are an important feature in speech, translation,
generation, IR, and other technologies. More training data and less
pruning generally lead to higher quality, but RAM is a limiting factor.
Further, systems consult language models so frequently that lookups
dominate CPU time.

This talk presents language modeling code with several optimizations to
improve time and space performance. Storing backoff information in
feature state reduces redundant lookups. Constructing known
distributions and biasing binary search speeds search and reduces page
faults. Memory mapping reduces load time. Bit level packing increases
locality. Stronger filtering removes n-grams that cannot be assembled
during decoding due to phrase and sentence constraints. The code is
currently integrated into Moses and being integrated into cdec and
Joshua. I will cover how my code works and how to use it in other
decoders.

Choosing the Right Evaluation for Machine Translation

2010-10-06T12:15:00.001-07:00

Time: Noon on Tuesday, October 12
Place: GHC 6501 (usual location)

Title: Choosing the Right Evaluation for Machine Translation: an Examination of Annotator and Automatic Metric Performance on Human Judgment Tasks
Authors: Michael Denkowski and Alon Lavie

Abstract:
This work examines the motivation, design, and practical results of several types of human evaluation tasks for machine translation. In addition to considering annotator performance and task informativeness over multiple evaluations, we explore the practicality of tuning automatic evaluation metrics to each judgment type in a comprehensive experiment using the METEOR metric. We present results showing clear advantages of tuning to certain types of judgments and discuss causes of inconsistency when tuning to various judgment data, as well as sources of difficulty in the human evaluation tasks themselves.

This work will be presented at AMTA 2010.

Models for Synchronous Grammar Induction for Statistical Machine Translation

2010-09-13T14:51:00.000-07:00

Title: Models for Synchronous Grammar Induction for Statistical Machine Translation

Presenters: Chris Dyer, LTI& Desai Chen, CSD undergraduate

Tuesday, September 14, at Noon to 1:30pm in GHC 6501.

Abstract: The last decade of research in Statistical Machine Translation (SMT) has seen rapid progress. The most successful methods have been based on synchronous context free grammars (SCFGs), which encode translational equivalences and license reordering between tokens in the source and target languages. Yet, while closely related language pairs can be translated with a high degree of precision now, the result for distant pairs is far from acceptable. In theory, however, the "right"' SCFG is capable of handling most, if not all, structurally divergent language pairs. This talk will report on the results of the 2010 Language Engineering Workshop held at Johns Hopkins University that the goal to focus on the crucial practical aspects of acquiring such SCFGs from bilingual, but otherwise unannotated, text. We started with existing algorithms for inducing unlabeled SCFGs (e.g. the popular Hiero model) and then used unsupervised learning methods to refine the syntactic constituents used in the translation rules of the grammar.

Syntax-to-Morphology Mapping in Factored Phrase-Based Statistical Machine Translation from English to Turkish

2010-06-28T13:50:00.000-07:00

Tuesday, June 29 at Noon, in GHC 6501

Title: Syntax-to-Morphology Mapping in Factored Phrase-Based Statistical
Machine Translation from English to Turkish

Authors: Reyyan Yeniterzi and Kemal Oﬂazer

Abstract:

We present a novel scheme to apply factored phrase-based SMT to a
language pair with very disparate morphological structures. Our
approach relies on syntactic analysis on the source side (English) and
then encodes a wide variety of local and non-local syntactic
structures as complex structural tags which appear as additional
factors in the training data. On the target side (Turkish), we only
perform morphological analysis and disambiguation but treat the
complete complex morphological tag as a factor, instead of separating
morphemes. We incrementally explore capturing various syntactic
substructures as complex tags on the English side, and evaluate how
our translations improve in BLEU scores. Our maximal set of source and
target side transformations, coupled with some additional techniques,
provide an 39\% relative improvement from a baseline 17.08 to 23.78
BLEU, all averaged over 10 training and test sets. Now that the
syntactic analysis on the English side is available, we also
experiment with more long distance constituent reordering to bring the
English constituent order close to Turkish, but ﬁnd that these
transformations do not provide any additional consistent tangible
gains when averaged over the 10 sets.

Chunk-Based EBMT

2010-05-12T13:52:00.000-07:00

When: noon on May 18
Where: GHC 6501

Speaker: Jaedy Kim

Topic: Chunk-Based EBMT

Abstract: Corpus driven machine translation approaches such as Phrase-Based Statistical Machine Translation and Example-Based Machine Translation have been successful by using word alignment to find translation fragments for matched source parts in a bilingual training corpus.
However, they still cannot properly deal with systematic translation for insertion or deletion words between two distant languages.
In this work, we used syntactic chunks as translation units to alleviate this problem, improve alignments and show improvement in BLEU for Korean to English and Chinese to English translation tasks.

Generalized templates for EBMT

2010-04-12T15:26:00.000-07:00

Speaker: Rashmi Gangadharaiah
Location: GHC 6501

Topic: Generalized templates for EBMT

Abstract:
-----------
Example-Based Machine Translation (EBMT), like other corpus based methods, requires substantial parallel training data. One way to reduce data requirements and improve translation quality is to generalize parts of the parallel corpus into translation templates. This automated generalization process requires clustering. In most clustering approaches the optimal number of clusters (N) is found empirically on a development set which often takes several days. We introduce a spectral clustering framework that automatically estimates the optimal N and removes unstable oscillating points. The new framework produces significant improvements in low-resource EBMT settings for English-to-French (~1.4 BLEU points), English-to-Chinese (~1 BLEU point), and English-to-Haitian (~2 BLEU points). The translation quality with templates created using automatically and empirically found best N were almost the same. By discarding “incoherent” points, a further boost in translation scores is observed, even above the empirically found N.

Two talks

2010-03-15T09:27:00.001-07:00

(1) Greg Hanneman:
Title: The Stat-XFER Group Submission for WMT '10

Abstract:
Each year, the Workshop in Statistical Machine Translation collects state-of-the-art MT results for a variety of European language pairs via a shared translation task. In this talk, I will describe the CMU's Stat-XFER MT group submission to this year's WMT French--English track, our third submission to the WMT series, using the Joshua decoder. A large focus will be on new modeling decisions or system-building techniques that have changed from eariler submissions based on new research carried out in our group. I will also present some open questions facing builders of large-scale hierarchcial MT systems in general.

(2) Vamshi Ambati:
Title: Making sense of Crowd data for Machine Translation

Abstract:
Quality of crowd data is a common concern in crowd-sourcing approaches to data collection. When working with crowd data, the objectives are two-fold - maximizing the quality of data from non-experts, and minimizing the cost of annotation by pruning noisy annotators.
I will discuss our recent experiments in Machine Translation for selection of high quality crowd translations by explicitly modeling annotator reliability based on agreement with other submissions. I will also present some preliminary results in cost minimization and report their adaptation and feasibility to machine translation.

Nonparametric Word Segmentation for Machine Translation

2010-02-18T22:43:00.000-08:00

Speaker: Thuylinh Nguyen
Title: Nonparametric Word Segmentation for Machine Translation
Thursday 18 Feb 2010. 12-1:30pm in GHC 4405.

In this talk we present an unsupervised word segmentation for machine
translation. The model utilizes existing nonparametric monolingual
segmentations. The monolingual segmentation model and the bilingual word
alignment model are coupled so that source text segmentation optimizes
the one-to-one mapping with the target text. Often, there are words in
the source language that do not appear in target language and vise
versa. Our model therefore models source language word deletion and word
insertion. The experiments show improvements on Arabic-English and
Chinese-English translation tasks.

LoonyBin: Making Empirical MT Reproducible, Efficient, and Less Annoying

2010-01-13T20:16:00.001-08:00

Speaker: Jonathan Clark
When: Tuesday, January 19 at Noon
Where: GHC 6501
What: Free Knowledge and Free Food
Title: LoonyBin: Making Empirical MT Reproducible, Efficient, and
Less Annoying

Abstract: Construction of machine translation systems has evolved into
a multi-stage workﬂow involving many complicated dependencies. Many
decoder distributions have addressed this by including monolithic
training scripts – train-factored-model.pl for Moses and mr_runmer.pl
for SAMT. However, such scripts can be tricky to modify for novel
experiments and typically have limited support for the variety of job
schedulers found on academic and commercial computer clusters. Further
complicating these systems are hyperparameters, which often cannot be
directly optimized by conventional methods requiring users to
determine which combination of values is best via trial and error. The
recently-released LoonyBin open-source workﬂow management tool
addresses these issues by providing: 1) a visual interface for the
user to create and modify workﬂows; 2) a well-deﬁned logging
mechanism; 3) a script generator that compiles visual workﬂows into
shell scripts, and 4) the concept of Hyperworkﬂows, which intuitively
and succinctly encodes small experimental variations within a larger
workﬂow. We also describe the Machine Translation Toolpack for
LoonyBin, which exposes state-of-the-art machine translation tools as
drag-and-drop components within LoonyBin.

MEMT and METEOR

2009-12-09T14:39:00.000-08:00

Kenneth Heafield and Michael Denkowski: Features for System Combination
(This is work done as an MT lab project.)

Michael will give an update on his recent work on the METEOR MT evaluation matrix.

10 Dec 2009, Thursday, 12:00-1:30, in GHC 6501

Lori's talk

2009-11-09T18:24:00.000-08:00

Speaker: Lori Levin
Where: GHC 6501
When: Nov 09, 2009 - Tuesday - Noon
Title: A Pendulum Swung Too Far
Abstract: This paper by Ken Church deals with the never ending battle between Empiricism and Rationalism,
esp. its incarnation in NLP. Lori will summarize and present the arguments formulated in the
paper. She will then continue with her own views on why linguistics
needs to be brought back into NLP and MT in particular.

Two talks

2009-08-10T12:52:00.000-07:00

Talk 1:
Nguyen Bach: Source-side Dependency Tree Reordering Models with Subtree Movements and Constraints

Abstract: We propose a novel source-side dependency tree reordering model for statistical machine translation, in which subtree movements and constraints are represented as reordering events associated with the widely used lexicalized reordering models. This model allows us to not only efficiently capture the statistical distribution of the subtree-to-subtree transitions in training data, but also utilize it directly at the decoding time to guide the search process. Using subtree movements and constraints as features in a log-linear model, we are able to help the reordering models make better selections. It also allows the subtle importance of monolingual syntactic movements to be learned alongside other reordering features. We show improvements in translation quality in English-Spanish and English-Iraqi translation tasks.

This is joint work with Qin Gao and Stephan Vogel.

Talk 2:
Francisco (Paco) Guzman: Reassessment of the Role of Phrase Extraction in SMT

Abstract: In this paper we study in detail the relation between word alignment and phrase extraction. First, we analyze different word alignments according to several characteristics and compare them to hand-aligned data. Then, we analyze the phrase-pairs generated by these alignments. We observed that the number of unaligned words has a large impact on the characteristics of the phrase table. A manual evaluation of phrase pair quality showed that the increase in the number of unaligned words results in a lower quality. Finally, we present translation results from using the number of unaligned words as features from which we obtain up to 2BP of improvement.

This is joint work with Qin Gao and Stephan Vogel.

Making Disfluent Output Slightly Less So: MT System Combination Search Spaces and Optimization

2009-06-15T08:55:00.000-07:00

Speaker: Kenneth Heafield

Title: Making Disfluent Output Slightly Less So:
MT System Combination Search Spaces and Optimization

Abstract: System combination merges several machine translation outputs
into a single improved sentence. This talk starts by summarizing the
approach including, a search space derived from the alignments, and
hypothesis scoring. The current search space focuses on picking words
in a roughly word synchronous way. Another search space under development
builds a directed graph in which aligned words correspond to a vertex and
each bigram corresponds to a directed edge. Search is conducted much like
a left-to-right MT decoder. Speed optimizations, which allow decoding at
5.5 sentences per second, apply to other MT systems in the areas of
duplicate handling, language model state, and multithreading. This speed
allows me to find hyperparameters by searching hundreds of parameter
combinations, each with a full round of tuning. In preparation for
last Friday's NIST submission, system combination improved 2.4 BLEU
points over the best component system for Urdu to English translation.

EBMT with external word alignment and chunk alignment

2009-04-28T16:38:00.000-07:00

Title: EBMT with external word alignment and chunk alignment.

Who: Jae Dong Kim
When: Tuesday May 12, 12:00pm
Where: NSH 3305

Abstract: Since both EBMT and SMT are data driven methods, more accurate word alignment improves system performance in EBMT as in SMT. However, EBMT has focused on finding analogous examples while SMT has achieved plausibly accurate word alignment. For this reason, it is natural that one thinks that EBMT can benefit from using SMT word alignment. In this talk, I am going to talk about our approach to make use of more accurate external word alignment from SMT in our EBMT system. I am also going to talk about my preliminary results with chunk alignment for translation in EBMT.

Language Model Adaptation for Difficult to Translate Phrases

2009-04-13T08:12:00.000-07:00

Presenter: Behrang Mohit
Title: Language Model Adaptation for Difficult to Translate Phrases
Date: Tuesday 12:30pm, 14 April 2009

Abstract:
We investigate the idea of adapting language models for phrases that
have poor translation quality. We apply a selective adaptation
criterion which uses a classifier to locate the most difficult phrase
of each source language sentence. A special adapted language model is
constructed for the highlighted phrase. Our adaptation heuristic uses
lexical features of the phrase to locate the relevant parts of the
parallel corpus for language model training. As we vary the
experimental setup by changing the size of the SMT training data, our
adaptation method consistently shows strong improvements over the
baseline systems.
This is a joint work with Frank Liberato and Rebecca Hwa.

Moving Beyond Phrase-Pairs: Dynamically Scoring Collections of Translation Examples

2009-03-12T09:06:00.001-07:00

Moving Beyond Phrase-Pairs: Dynamically Scoring Collections of
Translation Examples

Who: Aaron B. Phillips
When: Friday Mar 13, 12:00pm
Where: NSH 1507

Abstract:

Statistical Machine Translation has prospered because it is based on
models that are consistent and straightforward to optimize. The
log-linear model in particular allows the researcher to exploit
numerous, possibly dependent, features. However, the modeling approach
taken by SMT enforces a particular top-down view of the data using
phrase-pairs that does not easily allow for the integration of features
that may change from example to example. What I propose is a shift in
how the model is built. Inspired by Example-Based Machine Translation, I
calculate features for each example separately, but like SMT this
information is collected into a single log-linear model that is
straightforward to optimize. This is accomplished by identifying at
run-time the most appropriate collection of translation examples instead
of using precomputed phrase-pairs. A search is performed over each
example-specific feature such as the alignment quality, genre, or
context to determine a collection that maximizes the score. The weights
for each example-specific feature are adjustable during optimization and
allow for a trade-off between forming collections over all the examples
and forming collections that consist of a few high-quality examples.
This framework seeks to unify the approaches of EBMT and SMT. It results
in a model that is straightforward to optimize *and* allows the
integration of novel example-specific features.

An Overview of Tree-to-String Translation Models: Yang Liu

2009-02-03T10:27:00.000-08:00

Speaker: Yang Liu
Title: An Overview of Tree-to-String Translation Models

Abstract:

Recent research on statistical machine translation has lead to the rapid development of syntax-based translation models, in which syntactic information can be exploited to direct translation. In this talk, I will give an overview of tree-to-string translation models, one of the state-of-the-art syntax-based models. In a tree-to-string model, the source side is a phrase structure parse tree and the target side is a string. This talk includes the following topics: (1) naive tree-to-string model, (2) tree-sequence based tree-to-string model, (3) context-aware tree-to-string model, and (4) forest-based tree-to-string model. Experimental results show that forest-based tree-to-string model outperforms hierarchical phrase-based model significantly.

Short Bio:

Yang Liu is an Assistant Researcher at Institute of Computing Technology, Chinese Academy of Sciences. He graduated in Computer Science from Wuhan University in 2002. He received his PhD degree in Computer Science from Institute of Computing Technology, Chinese Academy of Sciences. His major research interests include statistical machine translation and Chinese information processing. His publications on discriminative word alignment and tree-to-string models have received wide attention. He served as PC member/Reviewer for TALIP, ACL, EMNLP, AMTA, and SSST.

Parallel Treebanks in Machine Translation

2009-01-13T08:21:00.000-08:00

Title: Parallel Treebanks in Machine Translation
Speaker: John Tinsley, Ph.D. student at the National Centre for Language Technology in DCU

Fast MT Pipeline: Introduction to tools you can use

2008-12-08T13:36:00.000-08:00

Date: 09-Dec-2008

Qin and Alok will report on their work to speed up some
of the MT processing by using parallel processing, with
an emphasis on the tools they have developed, to do this
kind of work.

Title: Fast MT Pipeline: Introduction to tools you can use.

Abstract: In this talk, we would like to introduce you to some
recently developed tools available for you to use, in order to
speed up the MT pipeline. The tools of focus are:
(i) multi-threaded giza: faster word alignment.
(ii) chaksi: Phrase-Extraction on the M45 cluster.
(iii) trambo: Decoding/MERT on the M45 cluster.

Presentations

2008-11-16T00:13:00.000-08:00

Date: 11 Nov 2008
Time: 12-1:30
Room: NSH 3305

Presentations:

Andreas Zollmann: Wider Pipelines: N-Best Alignments and Parses in MT Training

State-of-the-art statistical machine translation systems use hypotheses from several maximum a posteriori inference steps, including word alignments and parse trees, to identify translational structure and estimate the parameters of translation models. While this approach leads to a modular pipeline of independently developed components, errors made in these “single-best” hypotheses can propagate to downstream estimation steps that treat these inputs as clean, trustworthy training data. In this work we integrate N-best alignments and parses by using a probability distribution over these alternatives to generate posterior fractional counts for use in downstream estimation. Using these fractional counts in a DOPinspired syntax-based translation system, we show significant improvements in translation quality over a single-best trained baseline.

Silja Hildebrand: Combination of Machine Translation Systems via Hypothesis Selection from Combined N-Best Lists

Different approaches in machine translation achieve similar translation quality with a variety of translations in the output. Recently it has been shown, that it is possible to leverage the individual strengths of various systems and improve the overall translation quality by combining translation outputs. In this paper we present a method of hypothesis selection which is relatively simple compared to system combination methods which construct a synthesis of the input hypotheses. Our method uses information from n-best lists from several MT systems and features on the sentence level which are independent from the MT systems involved to improve the translation quality.

Bilingual-LSA based adaptation for statistical machine translation

2008-09-08T13:20:00.001-07:00

Date: 9 Sept 2008
Speaker: Wilson Tam
TITLE: Bilingual-LSA based adaptation for statistical machine translation

ABSTRACT:
We propose a novel approach to crosslingual language model (LM) and translation lexicon adaptation for statistical machine translation based on bilingual Latent Semantic Analysis (bLSA). bLSA enables latent topic distributions to be efficiently transferred across languages by enforcing a one-to-one topic correspondence during training. Using the proposed bLSA framework, model adaptation can be performed by, first, inferring the topic posterior distribution of the source text and then applying the inferred distribution to an N-gram LM of the target language and translation lexicon via marginal adaptation. The background phrase table is then enhanced with the additional phrase scores computed using the adapted translation lexicon.

The proposed framework also features rapid bootstrapping of LSA models for new languages based on a source LSA model of another language. Our approach was evaluated on the Chinese-to-English MT06 test set. Improvement in BLEU was observed when the adapted LM and the adapted translation lexicon were applied individually. When the adapted LM and the adapted lexicon were applied simultaneously, the gain in BLEU was additive yielding 28.91% in BLEU which is statistically significant at the 95% confidence interval with respect to the unadapted baseline with 28.06% in BLEU.

Learning from Human Interpreter Speech

2008-08-12T10:42:00.000-07:00

Speaker: Matthias Paulik

Title: "Learning from Human Interpreter Speech"
Date: 12 August 2008

Abstract:
Can spoken language translation (SLT) profit from human interpreter speech? In this talk, we explore scenarios which involve live human interpretation, off-line transcription and off-line translation on a massive scale. We consider the deployment of machine translation (MT) and automatic speech recognition (ASR) for the off-line transcription and translation tasks; our systems are trained on 80+ hours of audio data and on parallel text corpora of ~40 million words. To improve performance, we use the available human interpreter speech as an auxiliary information source to bias ASR and MT language models. We evaluate this approach on European Parliament Plenary Session (EPPS) data in three languages (English, Spanish and German), and report preliminary improvements in translation and transcription performance.

Improving Lexical Coverage of Syntax-driven MT by Re-structuring Non-isomorphic Trees

2008-07-15T10:40:00.000-07:00

Speaker: Vamshi Ambati
Date: 15 July 2008

Abtract:
Syntax-based approaches to statistical MT require syntax-aware methods for acquiring their underlying translation models from parallel data. This acquisition process can be driven by syntactic trees for either the source or target language, or by trees on both sides. Work to date has demonstrated that using trees for both sides suffers from severe coverage problems. Approaches that project from trees on one side, on the other hand, have higher levels of recall, but suffer from lower precision, due to the lack of syntactically-aware word alignments.

In this talk I first discuss extraction and the lexical coverage of the translation models learned in both of these scenarios. We will specifically look at how the non-isomorphic nature of the parse trees for the two languages effects recall and coverage. I will then discuss a novel technique for restructuring target parse trees, that generates highly isomorphic target trees that preserve the syntactic boundaries of constituents that were aligned in the original parse trees. I will conclude by discussing some experimental evaluation with an English-French MT System.

Scalable Decoding for Syntax based SMT

2008-06-10T10:59:00.000-07:00

Title: Scalable Decoding for Syntax based SMT

Speaker: Abhaya Agarwal
Joint work with Vamshi Ambati

Date: June 10, 2008
Time: 12am - 1:30pm
Location: NSH 3305

Abstract:
Most large scale SMT systems these days use a hybrid model, combining a handful of generative models in a discriminative log-linear framework whose weights are trained on a small amount of development data. Attempts to train fully discriminative MT models that use large number of features, face difficult scalability challenges because it requires repeated decoding of a large training set. In this talk, I will describe a decoder that aims to be efficient enough to allow such training and some initial experiments on training with large number of features.

Statistical Transfer MT Systems for French and German

2008-05-13T14:36:00.000-07:00

Speaker: Greg Hanneman

Date: Tuesday, May 13, at noon
Location: Wean Hall 4623

Title: Statistical Transfer MT Systems for French and German

Abstract: The AVENUE research group's statistical transfer system is a general framework for creating hybrid machine translation systems. It uses two main resources: a weighted synchronous context-free grammar, and a probabilistic bilingual lexicon of syntax-based word- and phrase-level translations. Over the last six months, we have developed new methods for extracting these resources automatically from parsed and word-aligned parallel corpora. In this talk, I will describe the resource-extraction process as it was applied to new French--English and German--English systems for the upcoming ACL workshop on statistical machine translation. Preliminary evaluation results --- both automatic and human-assessed --- will also be reviewed.

Simulating Sentence Pairs Sampling Process via Source and Target Language Models

2008-04-22T10:56:00.000-07:00

Speaker: Ngyuen Bach

Abstract: In a traditional word alignment process, each sentence pair is equally assigned an occurrence number, which is normalized during the training to produce the empirical probability. However, some sentences could be more valuable, reliable and appropriate than others. These sentences should therefore have a higher weight in the training. To solve this problem, we explored methods of resampling sentence pairs. We investigated three sets of features: sentence pair confidence (/sc/), genre-dependent sentence pair confidence (/gdsc/) and sentence-dependent phrase alignment confidence (/sdpc/) scores. These features were calculated over an entire training corpus and could easily be integrated into the phrase-based machine translation system.

Communicating Unknown Words in Machine Translation

2008-03-19T11:04:00.000-07:00

Speaker: Matthias Eck

Title: Communicating Unknown Words in Machine Translation

Abstract:
Unknown words are a major problem for every machine translation system. Regular evaluations and demos do not always show this very well, but in actual communication the lack of specialty vocabulary and named entity translations can seriously affect the communication ability.

A new approach is presented that uses monolingual encyclopedias and dictionaries to "communicate" unknown words. Instead of the actual unknown word, its definition is extracted and translated, which leads to considerable improvements in translation quality.

Trees that can help

2007-11-13T11:07:00.000-08:00

Speaker: Alok Parlikar

Title: (S (NP (NP Trees) (SBAR (WHNP that) (S (VP can)))) (VP help))

Summary:

For the past two months, I have been working with Alon Lavie and Stephan
Vogel, on Chinese and English parse-trees, to investigate answers to the
following questions:

(a) Can constituency information and word level alignments be used to
align nodes in trees of parallel sentences? How precisely matched
(in meaning) are the yields of these aligned nodes?
(b) Can the parse trees and word-level alignments be used for learning
reordering rules? If we use these rules to reorder source sentences,
can we do any better at translation?

The current results show that:

(a) - Node Alignments from hand-aligned data are very precise.
- Using automatic word-alignments to align nodes gives over 70%
precision and over 40% recall.
(b) Using a 10-best reordering of words in the source sentences, with
a "dumb" reordering strategy has shown a 0.005 improvement in BLEU
score.

I would like to talk about the approaches that we have taken here, and to
discuss about strategies for improving these results.

Sub-Phrasal Matching and Structural Templates in Example-Based MT

2007-10-09T11:09:00.000-07:00

Speaker: Aaron Phillips

Title: Sub-Phrasal Matching and Structural Templates in Example-Based MT

Example-Based Machine Translation (EBMT) encompasses many different
approaches to data-driven MT. In this work I first look at two different
paradigms of EBMT. I then combine the strengths of these two systems and
build a new engine that combines sub-phrasal matching with structural
templates. The end result is a melding of ideas from EBMT, SMT, and
Xfer. This synthesis results in higher translation quality and more
graceful degradation, yielding 1.5% to 7.5% relative improvement in BLEU
scores.

This work was recently presented at TMI. The full paper can be found
here:

http://dustoftheground.net/techne/research/sub-phrasal_matching_and_structural_templates_in_example-based_mt.pdf

Experiments with a Noun-Phrase driven Statistical Machine Translation System

2007-08-14T13:41:00.000-07:00

Title: Experiments with a Noun-Phrase driven Statistical Machine
Translation System

Speaker: Sanjika Hewavitharana
Date: Aug 14, 2007
Time: 12:00pm
Place: NSH 3305 [During MT Group Monthly Lunch]

Abstract:

Hierarchical translation models that use phrases with words as well as
sub-phrases have shown better performance than standard phrase based
systems. In this talk I will present a noun-phrase (NP) driven
statistical machine translation system. Using noun-phrases as the
decomposition unit, we build a two-level hierarchy of phrases. We first
identify noun-phrases in the data and replace them with a tag to produce
an NP tagged corpus. This corpus is then used to extract NP-tagged
phrase translation pairs. Both noun-phrases and NP-tagged phrases are
used in a two-level translation decoder. The two-level system shows
significant improvements over a baseline phrase-based SMT system. It
also produces longer matching phrases due to the generalization
introduced by tagging noun-phrases.

In Search of Better MT Evaluation Metric : Some Experiments

2007-05-22T13:45:00.000-07:00

Date: May 22, 2007
Presenter: Abhaya Agarwal
Title: In Search of Better MT Evaluation Metric : Some Experiments

Abstract: Area of automatic metrics for MT Evaluation has seen a lot of
activity in last 4-5 years. Starting with BLEU, many such metrics have
been proposed over the years including METEOR, HTER and ROUGE. While
these metrics achieve good correlation with human judgments at the
system level, situation remains bleak at individual sentence level. In
this talk, I will talk about some work that we have been doing towards
developing metrics with improved correlation with human judgments at the
sentence level. I will present the current results and some possible
future directions.

An Assessment of Language Elicitation without the Supervision of a Linguist

2007-04-17T13:46:00.000-07:00

Date: Apr 17, 2007
Presenter: Lori Levin and Alison Alvarez
Title: An Assessment of Language Elicitation without the Supervision of a Linguist

We created an elicitation corpus designed to elicit the morphosyntactic
features of a target language without the supervision of a linguist.
The corpus is composed of approximately 3200 English source sentences
that are then translated by a native speaker into the target language.
The design of our corpus was driven by our need to elicit morphosyntactic
language features without the supervision of a linguist. In a previous
paper we reported on a reverse Treebank and that was a deep
morphosyntactic tree with two parallel human language sentences. The
first is provided by reverse annotation and the second is acquired
through elicitation. This presentation will focus on the extent to which
we able to acquire our morphosyntactic information from our translated
corpora and the types of errors we encountered, both from the perspective
of the translator and the corpus itself.

Translation Model Pruning

2007-02-20T14:03:00.000-08:00

Date: Feb 20, 2007

Presenter: Matthias Eck

Title: Translation Model Pruning

SALM: Suffix Array and its Applications in Empirical Language Processing

2007-01-16T14:05:00.000-08:00

Date: Jan 16, 2007

Presenter: Ying(Joy) Zhang

Title: SALM: Suffix Array and its Applications in Empirical Language Processing

Simulating Multiple Translations and ASR Transcripts for Applications in Multilingual Spoken Document Classification

2006-11-21T15:36:00.000-08:00

Title: Simulating Multiple Translations and ASR Transcripts for Applications in Multilingual Spoken Document Classification

Speaker: Wei-Hao Lin from the Informedia group

Abstract:
We propose a statistical model to simulate multiple documents and
their translations (e.g. Chinese documents and their English
translations), and apply the model in the task of classifying
multilingual documents. The model, based on a frequency matching
principle, predicts that previous approaches to building classifiers
from a common language (e.g., English) are not optimal for
multilingual collections with unbalanced numbers of documents, and a
proposed multilingual representation can outperform the mono-lingual
bag-of-words representation. We also investigate the possibility of
combining multiple ASR transcripts and translations through
re-weighting. The validity of our model is strongly supported by
the close match between predictions of the simulation model and the
empirical results of classifying multilingual spoken documents from
broadcast news in three languages.

Coupling of ASR+MT: Initial Experiments & Future Directions

2006-10-17T15:08:00.000-07:00

Speaker: Ian Lane

Title: Tighter Coupling of ASR+MT: Initial Experiments & Future Directions

Abstract:
In this talk, I will first give a brief overview of my PhD work entitled "Flexible Spoken Language Understanding based on Topic Classification and Domain Detection", and describe how the proposed approaches can be applied to applications other than speech-to-speech translation. I will
then describe my current work which focuses on improving coupling between ASR and Machine-Translation Systems, specifically, when applied to conversational speech. Finally, I will propose future directions for which I hope to receive a large amount of feedback.

Anchor-Based Symmetric Probabilistic Alignment

2006-05-23T15:18:00.000-07:00

Date: May 23, 2006

Presenter: Jae Dong Kim,

Title: Anchor-Based Symmetric Probabilistic Alignment

Can the Internet help improve Machine Translation?

2006-04-18T15:19:00.000-07:00

Date: April 18, 2006

Presenter: Ari Font-Llitjos

Title: Can the Internet help improve Machine Translation?

The MILE Corpus for Less Commonly Taught Languages

2006-02-21T15:21:00.000-08:00

Date: February 21, 2006

Presenter: Alison Alvarez,

Title: The MILE (Minor Language Elicitation) Corpus for Less Commonly Taught Languages

Generalization and Context-Sensitivity for Example-Based Machine Translation

2005-12-08T15:22:00.000-08:00

Date: December 08, 2005

Presenter: Ralf Brown

Title: Generalization and Context-Sensitivity for Example-Based Machine Translation

Speech Translation with Multiple Speech and Translation Hypotheses

2005-11-15T15:24:00.000-08:00

Date: November 15, 2005

Presenter: Chiori Hori

Title: Speech Translation with Multiple Speech and Translation Hypotheses

Adapting Resources for Parsing Arabic Dialects

2005-10-18T15:25:00.000-07:00

Date: October 18, 2005

Presenters: Rebecca Hwa and Carol Nichols from the NLP Lab at Pitt

Title:Adapting Resources for Parsing Arabic Dialects

Named Entity Extraction and Translation from Multimedia Documents

2005-09-13T15:26:00.001-07:00

Date: September 13, 2005.

Presenter: Fei Huang,

Title: "Named Entity Extraction and Translation from Multimedia Documents".

David Chiang's ACL-2005 paper

2005-08-23T15:28:00.000-07:00

Date: August 23, 2005.

Presenter: Joy (Ying Zhang)

Title: "David Chiang's ACL-2005 paper : A Hierarchical Phrase-Based Model for Statistical Machine Translation".

Semi-Automated Elicitation Corpus Generation

2005-06-21T15:29:00.000-07:00

Date: June 21, 2005.

Presenter: Alison Alvarez,

Title: "Semi-Automated Elicitation Corpus Generation".

Using Collocations to Assess MT Quality

2005-05-10T15:30:00.000-07:00

Date: May 10, 2005.

Presenter: Benjamin Han,

Title: Using Collocations to Assess MT Quality

Mining Translations for Key Phrases from Web Corpora

2005-04-19T15:31:00.000-07:00

Date: April 19, 2005.

Presenters: Joy (Ying Zhang) and Fei Huang,

Title: "Mining Translations for Key Phrases from Web Corpora".

A New Approach for Multi-Engine MT

2005-03-22T15:31:00.000-08:00

Date: March 22, 2005.

Presenter: Shyamsundar Jayaraman,

Title: "A New Approach for Multi-Engine MT" (MEMT)

The METEOR metric

2005-02-22T15:33:00.000-08:00

February 22, 2005.

Presenter: Alon Lavie,

Title: "The METEOR metric" (MT Evaluation)