Marzieh Fadaee - 2017 - Data Augmentation for Low-Resource Neural Machine Translation

•

0 likes•82 views

Association for Computational Linguistics

This document presents an approach called Translation Data Augmentation (TDA) to improve neural machine translation, especially for low-resource languages. TDA generates additional training sentences by altering existing parallel sentences to target rare words. This increases the size and diversity of the training data. Experimental results show that training with augmented data significantly improves BLEU scores over the baseline, with gains of up to 2.9 BLEU points for a simulated low-resource English-German translation task. TDA particularly helps with translating low-frequency words, which neural models often handle inaccurately due to lack of data.

Conclusions
Data Augmentation for Low-Resource
Neural Machine Translation
Summary
Marzieh Fadaee Arianna Bisazza Christof Monz
Informatics Institute, University of Amsterdam
Approach: Translation Data Augmentation (TDA)
55th Annual Meeting of the Association for Computational Linguistics (ACL 2017), Vancouver, Canada m.fadaee@uva.nl
Rare Translation Generation (DE➞EN)
NMT Results (BLEU)
Data Augmentation
Words in
Words in
Words in
generated during translation
not generated during translation
aﬀected by augmentation
Vrare Vreference
Vrare Vreference
Vrare Vreference
,
,
,
,
,
,
,
,
,
,
,
,
The generated translation length to reference length ratio is on average 7% higher
Model testset2014 testset2015 testset2016
TDAr=1
TDAr 1
Baseline
Back-trans1:1
Back-trans1:1
Altering only one word per sentence
Closest work is back-translation of monolingual data
(Sennrich et al. ACL 2016)
Our approach focuses on non meaning-preserving
augmentation
… that is compatible
with the contextChoosing the best
translation of s0
i
LSTM
LSTM
…
I had been told that you would voluntarily be speaking today.
mir wurde signalisiert, sie würden heute freiwillig sprechen.
New sentence pair:
Altering one or multiple words per sentence
TDA signiﬁcantly improves translation quality
Substituting several rare words is preferable even
though the augmented sentences are likely to be noisier
DE➞EN
13.4 (+2.1)
I had been told that you would not be speaking today .
mir wurde signalisiert , sie würden heute nicht sprechen .
Model testset2014 testset2015 testset2016
11.4 (+0.8)
10.6
11.9 (+1.3)TDAr=1
TDAr 1 12.6 (+2.0)
Baseline
12.2 (+0.9)
11.3
13.7 (+2.4)
14.6 (+1.5)
13.1
15.2 (+2.1)
15.4 (+2.3)
9.0 (+0.8)
8.2
10.4 (+2.2)
10.7 (+2.5)
10.4 (+1.2)
9.2
11.2 (+2.0)
11.5 (+2.3)
12.0 (+1.0)
11.0
13.5 (+2.5)
13.9 (+2.9)
EN➞DE
1000 2000 3000
testset2014
testset2015
testset2016
Baseline
Baseline
Baseline
TDAr 1
TDAr 1
TDAr 1
“freiwillig”
t0
j = argmax
t2trans(s0
i)
Plex(t|s0
i)Plexinv
(s0
i|t)PLMT
(t|tj 1
1 )
LM probability
threshold
“voluntarily”
s0
i
-
-
- …
ungefragt
freiwillig
trans(s0
i)
Image Processing
Has not been done in Natural Language Processing
One possible approach is paraphrasing which is meaning-
preserving
Flipping, cropping, tilting, altering the RGB channels
Neural Machine Translation models perform best when an abundance of parallel
data is available
Acquiring human translations for low-resource language pairs is costly
Our approach
As a result training with the augmented bitext achieves signiﬁcant BLEU
improvements in a simulated low-resource English⬌German translation setting
alters existing parallel sentences targeting low-frequency words
augments the data by generating new diverse context for low-frequency
words and the corresponding translations
We present a data augmentation technique to enrich the training data targeting rare words
Increasing the size of the training data by diversifying the context of rare words yields better translations
Generation of correct rare words during translation increases
The attention scores of rare words are on average 8.8% higher than the baseline model
Hence translation of low-frequency words is difﬁcult and often inaccurate
t 2 trans(s0
i)
Vrare
Data
371K
731K
4.5M
6M
371K
731K
4.5M
6M
Data
Simulated low-resource MT setting
This research was funded in part by
the Netherlands Organization for
Scientiﬁc Research (NWO) under
project numbers 639.022.213 and
639.021.646

This document discusses sequence to sequence learning with Tensor2Tensor (T2T) and sequence models. It provides an overview of T2T, which is a library for deep learning models and datasets. It discusses basics of sequence models including recurrent neural networks (RNNs), convolutional models, and the Transformer model based on attention. It encourages experimenting with different sequence models and datasets in T2T.

Neural Text Embeddings for Information Retrieval (WSDM 2017)

Bhaskar Mitra

Language Grid

lindh

This document summarizes a workshop on the Language Grid, which is a service-oriented infrastructure for multilingual societies. It discusses how the Language Grid provides various language services, such as machine translation, dictionaries, and parallel texts. It also describes how these atomic services can be composed to create new multilingual applications and services. Finally, it outlines several research projects using the Language Grid, including analyzing machine translation-mediated communication, developing multilingual localization systems, and extending the Language Grid's capabilities.

Real-time DirectTranslation System for Sinhala and Tamil Languages.

Sheeyam Shellvacumar

Sequence to Sequence Learning with Neural Networks

Nguyen Quang

This document discusses sequence to sequence learning with neural networks. It summarizes a seminal paper that introduced a simple approach using LSTM neural networks to map sequences to sequences. The approach uses two LSTMs - an encoder LSTM to map the input sequence to a fixed-dimensional vector, and a decoder LSTM to map the vector back to the target sequence. The paper achieved state-of-the-art results on English to French machine translation, showing the potential of simple neural models for sequence learning tasks.

Pangeanic Cor-ActivaTM-Neural machine translation Taus Tokyo 2017

Manuel Herranz

The document discusses recent developments in machine translation technology, including the web, databases, and neural networks. It provides three key points: 1. New projects like Cor and ActivaTM have created more flexible machine translation systems that can translate documents and web content, and allow language assets to be stored and accessed from a single database. 2. Neural machine translation has surpassed statistical machine translation in some areas, obtaining better translation quality as measured by metrics like BLEU, TER, and WER, especially for medium-length sentences. 3. With more training data and continued improvements, neural machine translation may soon achieve human-level quality and become the predominant machine translation approach within a few years, potentially requiring

ANALYZING ARCHITECTURES FOR NEURAL MACHINE TRANSLATION USING LOW COMPUTATIONA...

ijnlc

With the recent developments in the field of Natural Language Processing, there has been a rise in the use of different architectures for Neural Machine Translation. Transformer architectures are used to achieve state-of-the-art accuracy, but they are very computationally expensive to train. Everyone cannot have such setups consisting of high-end GPUs and other resources. We train our models on low computational resources and investigate the results. As expected, transformers outperformed other architectures, but there were some surprising results. Transformers consisting of more encoders and decoders took more time to train but had fewer BLEU scores. LSTM performed well in the experiment and took comparatively less time to train than transformers, making it suitable to use in situations having time constraints.

ANALYZING ARCHITECTURES FOR NEURAL MACHINE TRANSLATION USING LOW COMPUTATIO...

kevig

This document discusses the integration of speech recognition with computer-assisted translation. It begins by introducing machine translation and computer-assisted translation, then describes how automatic speech recognition works and how it can be integrated with translation. Key approaches to integration include using word graphs from ASR and MT systems or rescoring ASR hypotheses with translation models. Neural machine translation models that use attention mechanisms are also discussed. The document concludes by noting areas for further development in reducing human effort in translation and increasing quality and effectiveness of speech-to-text and translation tools.

Language Technology Enhanced Learning

telss09

This document provides an overview of using latent semantic analysis (LSA) and the R programming language for language technology enhanced learning applications. It describes using LSA to create a semantic space to compare documents and evaluate student writings. It also demonstrates clustering terms based on their semantic similarity and visualizing networks in R. Evaluation results show LSA machine scores for essay quality had a Spearman's rank correlation of 0.687 with human scores, outperforming a pure vector space model.

MACHINE-DRIVEN TEXT ANALYSIS

Massimo Schenone

Natural language processing (NLP) involves developing systems that can process and understand human language. This document discusses NLP tools and techniques for representing text numerically so it can be analyzed by machine learning algorithms. It covers topics like tokenization, part-of-speech tagging, named entity recognition, vector space models, term frequency-inverse document frequency (TF-IDF) weighting, and word embeddings which represent words as dense vectors of numbers. Popular Python libraries for NLP and text analysis are also introduced.

Contemporary Models of Natural Language Processing

Katerina Vylomova

This document provides an overview of contemporary models in natural language processing, focusing on language modeling and machine translation. It introduces n-gram models, distributional semantics, and neural language models such as word2vec. For machine translation, it covers statistical machine translation using noisy channel modeling and alignment, and neural machine translation using sequence-to-sequence models with attention. Neural models generally outperform statistical models on machine translation tasks.

End-to-End Memory Networks with Knowledge Carryover for Multi-Turn Spoken Lan...

Yun-Nung (Vivian) Chen

The document describes an end-to-end memory network model for multi-turn spoken language understanding. The model encodes context from previous utterances using an attention mechanism over the memory of past utterances. It then performs slot tagging on the current utterance incorporating the contextual knowledge. Experiments on a Cortana dataset show the model outperforms alternatives, achieving 67.1% accuracy by encoding both history and current utterances with the memory network.

EMPLOYING PIVOT LANGUAGE TECHNIQUE THROUGH STATISTICAL AND NEURAL MACHINE TRA...

ijnlc

The quality of Neural Machine Translation (NMT) systems like Statistical Machine Translation (SMT) systems, heavily depends on the size of training data set, while for some pairs of languages, high-quality parallel data are poor resources. In order to respond to this low-resourced training data bottleneck reality, we employ the pivoting approach in both neural MT and statistical MT frameworks. During our experiments on the Persian-Spanish, taken as an under-resourced translation task, we discovered that, the aforementioned method, in both frameworks, significantly improves the translation quality in comparison to the standard direct translation approach.

Build your own ASR engine

Korakot Chaovavanich

This document provides an overview of building an automatic speech recognition (ASR) engine. It discusses speech as a natural modality with high throughput that needs to account for errors in ASR output. It describes the components of a dialogue system including the ASR, natural language understanding, text-to-speech, and a dialogue manager. The document then discusses the components inside the recognizer including the acoustic model, language model, feature extraction using MFCC, and decoding using techniques like beam search. It also discusses topics like building the lexicon, acoustic modeling, and using deep learning approaches in ASR.

Text Mining Infrastructure in R

Ashraf Uddin

R is a free software environment for statistical analysis and graphics. This document discusses using R for text mining, including preprocessing text data through transformations like stemming, stopword removal, and part-of-speech tagging. It also demonstrates building term document matrices and classifying text with k-nearest neighbors (KNN) algorithms. Specifically, it shows classifying speeches from Obama and Romney with over 90% accuracy using KNN classification in R.

Effect of Machine Translation in Interlingual Conversation: Lessons from a Fo...

Kotaro Hara

Our talk at CHI2015 in Seoul, South Korea. Find more information at www.kotarohara.com . YouTube: https://youtu.be/isqsYLkX9gA Makeability Lab: http://www.cs.umd.edu/~jonf/ Microsoft Research: http://research.microsoft.com/ ABSTRACT Language barrier is the primary challenge for effective cross-lingual conversations. Spoken language translation (SLT) is perceived as a cost-effective alternative to less affordable human interpreters, but little research has been done on how people interact with such technology. Using a prototype translator application, we performed a formative evaluation to elicit how people interact with the technology and adapt their conversation style. We conducted two sets of studies with a total of 23 pairs (46 participants). Participants worked on storytelling tasks to simulate natural conversations with 3 different interface settings. Our findings show that collocutors naturally adapt their style of speech production and comprehension to compensate for inadequacies in SLT. We conclude the paper with the design guidelines that emerged from the analysis.

Jérémy Ferrero - 2017 - Using Word Embedding for Cross-Language Plagiarism ...

Association for Computational Linguistics

The document presents research on using word embeddings for cross-language plagiarism detection. It introduces the task of cross-language plagiarism detection and evaluates several state-of-the-art methods. New methods proposed include CL-CTS-WE, CL-WES, and CL-WESS, which augment existing approaches with word embeddings. Experimental results show CL-WESS outperforms other state-of-the-art methods, and combining methods through decision tree fusion further improves performance.

Big Data Spain 2017 - Deriving Actionable Insights from High Volume Media St...

Apache OpenNLP

Media analysts have to deal with with analyzing high volumes of real-time news feeds and social media streams which is often a tedious process because they need to write search profiles for entities. Python tools like NLTK do not scale to large production data sets and cannot be plugged into a distributed scalable frameworks like Apache Flink. Apache Flink being a streaming first engine is ideally suited for ingesting multiple streams of news feeds, social media, blogs etc.. and for being able to do streaming analytics on the various feeds. Natural Language Processing tools like Apache OpenNLP can be plugged into Flink streaming pipelines so as to be able to perform common NLP tasks like Named Entity Recognition (NER), Chunking, and text classification. In this talk, we’ll be building a real-time media analyzer which does Named Entity Recognition (NER) on the individual incoming streams, calculates the co-occurrences of the named entities and aggregates them across multiple streams; index the results into a search engine and being able to query the results for actionable insights. We’ll also be showing as to how to handle multilingual documents for calculating co-occurrences. NLP practitioners will come away from this talk with a better understanding of how the various Apache OpenNLP components can help in processing large streams of data feeds and can easily be plugged into a highly scalable and distributed framework like Apache Flink.

70 C o m m u n i C at i o n s o f t h E a C m j u.docx

evonnehoggarth79783

70 C o m m u n i C at i o n s o f t h E a C m | j u ly 2 0 1 3 | v o l . 5 6 | n o . 7 contributed articles V o i c e i n p u t i s a major requirement for practical question answering (QA) systems designed for smartphones. Speech-recognition technologies are not fully practical, however, due to fundamental problems (such as a noisy environment, speaker diversity, and errors in speech). Here, we define the information distance between a speech-recognition result and a meaningful query from which we can reconstruct the intended query, implementing this framework in our RSVP system. In 12 test cases covering male, female, child, adult, native, and non-native English speakers, each with 57 to 300 questions from an independent test set of 300 questions, RSVP on average re- duced the number of errors by 16% for native speakers and by 30% for non- native speakers over the best-known speech-recognition software. The idea was then extended to translation in the QA domain. In our project, which is supported by Canada’s International Develop- ment Research Centre (http://www. idrc.ca/), we built a voice-enabled cross-language QA search engine for cellphone users in the developing world. Using voice input, a QA system would be a convenient tool for people who do not write, for people with im- paired vision, and for children who might wish their Talking Tom or R2- D2 really could talk. The quality of today’s speech-recognition technologies, exemplified by systems from Google, Microsoft, and Nuance does not fully meet such needs for several reasons: ˲ Noisy environments in common audio situations;1 ˲ Speech variations, as in, say, adults vs. children, native speakers vs. non-native speakers, and female vs. male, especially when individual voice-input training is not possible, as in our case; and ˲ Incorrect and incomplete sentences; even customized speech-recognition systems would fail due to coughing, breaks, corrections, and the inability to distinguish between, say, “sailfish” and “sale fish.” Speech-recognition systems can be trained for a “fixed command set” of up to 10,000 items, a paradigm that information Distance Between What i said and What it heard D o i : 1 0 . 1 1 4 5 / 2 4 8 3 8 5 2 . 2 4 8 3 8 6 9 The RSVP voice-recognition search engine improves speech recognition and translation accuracy in question answering. BY YanG tanG, Di WanG, JinG Bai, XiaoYan Zhu, anD minG Li key insights focusing on an infinite but highly structured domain (such as Qa), we significantly improve general-purpose speech recognition results and general-purpose translation results. assembling a large amount of internet data is key to helping us achieve these goals; in the highly structured Qa domain, we collected millions of human-asked questions covering 99% of question types. RsVP development is guided by a theory involving informatio.

Expressiveness, Simplicity and Users

greenwop

Craig Chambers discusses his past work on programming languages and compilers and his current work on Flume, a data-parallel programming system. Flume aims to make data-parallel programming easy by providing high-level abstractions while automatically optimizing and executing pipelines. Flume builds an execution graph from data-parallel operations and optimizes it into MapReduce jobs before execution. Early experience shows Flume is easier for users than raw MapReduce and the optimizer helps improve performance. Future work includes expanding Flume to more execution substrates and auto-tuning pipelines.

“Neural Machine Translation for low resource languages: Use case anglais - wo...

Paris Women in Machine Learning and Data Science

End-to-end Speech Recognition with Recurrent Neural Networks (D3L6 Deep Learn...

Universitat Politècnica de Catalunya

https://telecombcn-dl.github.io/2017-dlsl/ Winter School on Deep Learning for Speech and Language. UPC BarcelonaTech ETSETB TelecomBCN. The aim of this course is to train students in methods of deep learning for speech and language. Recurrent Neural Networks (RNN) will be presented and analyzed in detail to understand the potential of these state of the art tools for time series processing. Engineering tips and scalability issues will be addressed to solve tasks such as machine translation, speech recognition, speech synthesis or question answering. Hands-on sessions will provide development skills so that attendees can become competent in contemporary data analytics tools.

07-Effect-Of-Machine-Translation-In-Interlingual-Conversation.pdf

simonp16

The document summarizes research on the effect of machine translation on interlingual conversations. Researchers studied conversations between French-German and English-German language pairs using a translator tool with different settings (closed captions only, text-to-speech only, or both). Quantitative analysis found that participants preferred settings with closed captions and perceived conversation quality improved over multiple rounds. Content analysis revealed that errors in proper noun recognition, word order, and speech overlap caused issues, but participants also adapted their speech and were forgiving of errors. The research aims to understand how people interact with speech translation technology and derive design guidelines.

Natural Language Processing - Research and Application Trends

Shreyas Suresh Rao

Topic model an introduction

Yueshen Xu

IMPROVE THE QUALITY OF IMPORTANT SENTENCES FOR AUTOMATIC TEXT SUMMARIZATION

csandit

There are sixteen known methods for automatic text summarization. In our study we will use Natural language processing NLP within hybrid approach that will improve the quality of important sentences selection by thickening sentence score along with reducing the number of long sentences that would be included in the final summarization. The based approach which is used in the algorithm is Term Occurrences.

Muis - 2016 - Weak Semi-Markov CRFs for NP Chunking in Informal Text

Association for Computational Linguistics

This paper contributes a noun phrase-annotated SMS corpus and proposes a weak semi-Markov CRF model for noun phrase chunking in informal text. The weak semi-CRF model improves training speed over linear-CRF and semi-CRF models while maintaining similar accuracy. Experiments on the SMS corpus show the weak semi-CRF achieves F1 scores comparable to other models but trains faster, especially with larger training data sizes.

Castro - 2018 - A High Coverage Method for Automatic False Friends Detection ...

Association for Computational Linguistics

This document presents a new method for automatically detecting false friends between Spanish and Portuguese using word embeddings. The method builds word vector spaces for each language using word2vec, finds a linear transformation between the spaces, and measures vector distances to classify word pairs as cognates or false friends. In experiments on a dataset of 710 word pairs, the method achieved state-of-the-art accuracy of 77.28% and high coverage of 97.91%, outperforming previous work. Future work will explore using different word embeddings and fine-grained classifications of partial false friends.

Similar to Marzieh Fadaee - 2017 - Data Augmentation for Low-Resource Neural Machine Translation

ANALYZING ARCHITECTURES FOR NEURAL MACHINE TRANSLATION USING LOW COMPUTATIONA...

kevig

Integration of speech recognition with computer assisted translation

Chamani Shiranthika

Language Technology Enhanced Learning

telss09

MACHINE-DRIVEN TEXT ANALYSIS

Massimo Schenone

Contemporary Models of Natural Language Processing

Katerina Vylomova

End-to-End Memory Networks with Knowledge Carryover for Multi-Turn Spoken Lan...

Yun-Nung (Vivian) Chen

EMPLOYING PIVOT LANGUAGE TECHNIQUE THROUGH STATISTICAL AND NEURAL MACHINE TRA...

ijnlc

Build your own ASR engine

Korakot Chaovavanich

Text Mining Infrastructure in R

Ashraf Uddin

Effect of Machine Translation in Interlingual Conversation: Lessons from a Fo...

Kotaro Hara

Jérémy Ferrero - 2017 - Using Word Embedding for Cross-Language Plagiarism ...

Association for Computational Linguistics

Big Data Spain 2017 - Deriving Actionable Insights from High Volume Media St...

Apache OpenNLP

70 C o m m u n i C at i o n s o f t h E a C m j u.docx

evonnehoggarth79783

Expressiveness, Simplicity and Users

greenwop

“Neural Machine Translation for low resource languages: Use case anglais - wo...

Paris Women in Machine Learning and Data Science

End-to-end Speech Recognition with Recurrent Neural Networks (D3L6 Deep Learn...

Universitat Politècnica de Catalunya

07-Effect-Of-Machine-Translation-In-Interlingual-Conversation.pdf

simonp16

Natural Language Processing - Research and Application Trends

Shreyas Suresh Rao

Topic model an introduction

Yueshen Xu

IMPROVE THE QUALITY OF IMPORTANT SENTENCES FOR AUTOMATIC TEXT SUMMARIZATION

csandit

Similar to Marzieh Fadaee - 2017 - Data Augmentation for Low-Resource Neural Machine Translation (20)

ANALYZING ARCHITECTURES FOR NEURAL MACHINE TRANSLATION USING LOW COMPUTATIONA...

Integration of speech recognition with computer assisted translation

Language Technology Enhanced Learning

MACHINE-DRIVEN TEXT ANALYSIS

Contemporary Models of Natural Language Processing

End-to-End Memory Networks with Knowledge Carryover for Multi-Turn Spoken Lan...

EMPLOYING PIVOT LANGUAGE TECHNIQUE THROUGH STATISTICAL AND NEURAL MACHINE TRA...

Build your own ASR engine

Text Mining Infrastructure in R

Effect of Machine Translation in Interlingual Conversation: Lessons from a Fo...

Jérémy Ferrero - 2017 - Using Word Embedding for Cross-Language Plagiarism ...

Big Data Spain 2017 - Deriving Actionable Insights from High Volume Media St...

70 C o m m u n i C at i o n s o f t h E a C m j u.docx

Expressiveness, Simplicity and Users

“Neural Machine Translation for low resource languages: Use case anglais - wo...

End-to-end Speech Recognition with Recurrent Neural Networks (D3L6 Deep Learn...

07-Effect-Of-Machine-Translation-In-Interlingual-Conversation.pdf

Natural Language Processing - Research and Application Trends

Topic model an introduction

IMPROVE THE QUALITY OF IMPORTANT SENTENCES FOR AUTOMATIC TEXT SUMMARIZATION

More from Association for Computational Linguistics

Muis - 2016 - Weak Semi-Markov CRFs for NP Chunking in Informal Text

Association for Computational Linguistics

Castro - 2018 - A High Coverage Method for Automatic False Friends Detection ...

Association for Computational Linguistics

Castro - 2018 - A Crowd-Annotated Spanish Corpus for Humour Analysis

Association for Computational Linguistics

This document describes a Spanish language corpus for humor analysis that was created through crowd-sourcing annotations. Over 27,000 tweets were collected from humorous accounts and annotated through a web interface. The corpus contains over 100,000 annotations of the tweets' humor and funniness. Inter-annotator agreement was higher for this corpus than a previous Spanish humor corpus. The dataset will help analyze subjectivity in humor and was used in a shared task on humor classification and funniness prediction.

Muthu Kumar Chandrasekaran - 2018 - Countering Position Bias in Instructor In...

Association for Computational Linguistics

This document discusses position bias in instructor interventions in MOOC discussion forums. It finds that instructors are more likely to intervene in threads that appear higher on the discussion forum user interface due to their recent activity. To address this, it proposes a debiased classifier that weights examples based on their propensity for intervention. It finds this approach identifies intervention opportunities that were overlooked due to position bias. The debiased classifier outperforms a standard classifier on several metrics, demonstrating it can better predict unbiased intervention needs.

Daniel Gildea - 2018 - The ACL Anthology: Current State and Future Directions

Association for Computational Linguistics

The document summarizes the history and current state of the ACL Anthology, a repository of publications from ACL-sponsored conferences. It discusses how the Anthology was established in 2001 and is now maintained by volunteers, containing over 45,000 papers. The presentation calls for community involvement to help future-proof the Anthology through efforts like migrating its infrastructure and improving documentation. It also proposes hosting the Anthology on the main ACL website and recruiting a new editor.

Elior Sulem - 2018 - Semantic Structural Evaluation for Text Simplification

Association for Computational Linguistics

The document presents SAMSA, a new automatic evaluation measure for structural text simplification. SAMSA uses semantic parsing to measure the preservation of semantic structures and relations between an original text and its simplified version. It correlates significantly better with human judgments of meaning preservation and structural simplicity than prior reference-based metrics. SAMSA is the first evaluation method designed specifically for structural simplification operations like sentence splitting.

Daniel Gildea - 2018 - The ACL Anthology: Current State and Future Directions

Association for Computational Linguistics

Wenqiang Lei - 2018 - Sequicity: Simplifying Task-oriented Dialogue Systems w...

Association for Computational Linguistics

(1) Sequicity is a framework that simplifies task-oriented dialogue systems using single sequence-to-sequence architectures. (2) It formalizes dialogues as sequences of belief spans and responses and decodes them in two stages: generating a belief span followed by a response. (3) An experiment on two datasets found that a two-stage CopyNet instantiation of Sequicity outperformed several baselines in effectiveness, efficiency and handling out-of-vocabulary requests.

Matthew Marge - 2017 - Exploring Variation of Natural Human Commands to a Rob...

Association for Computational Linguistics

The document summarizes a study that explored how people's strategies for giving commands to a robot change over time during a collaborative navigation task. Ten participants each directed a robot for one hour via dialogue. Initially, participants predominantly used metric units like distances in their commands, but over time their commands increasingly referred to environmental landmarks. The study collected audio, text, and robot data to analyze parameters in commands. Future work aims to automate dialogue response generation based on this data.

Venkatesh Duppada - 2017 - SeerNet at EmoInt-2017: Tweet Emotion Intensity Es...

Association for Computational Linguistics

The document describes a system for estimating emotion intensity in tweets. It takes a lexicon-based and word vector-based approach to create sentence embeddings for tweets. Various regression models are trained and an ensemble is used to predict emotion intensity scores between 0-1 for anger, sadness, joy and fear. The system achieved third place in predicting emotion intensity and second place for intensities over 0.5. Future work involves using contextual sentence embeddings to improve predictions.

Satoshi Sonoh - 2015 - Toshiba MT System Description for the WAT2015 Workshop

Association for Computational Linguistics

This document describes Toshiba's machine translation system submitted to the WAT2015 workshop. It discusses using statistical post-editing (SPE) to improve rule-based machine translation (RBMT) output, as well as combining SPE and SMT systems using reranking with recurrent neural network language models. Experimental results show that the combined system achieved the best BLEU and RIBES scores compared to the individual SPE and SMT systems on several language pairs, including Japanese-English and Chinese-Japanese. However, human evaluation correlations were not entirely clear.

Chenchen Ding - 2015 - NICT at WAT 2015

Association for Computational Linguistics

John Richardson - 2015 - KyotoEBMT System Description for the 2nd Workshop on...

Association for Computational Linguistics

The document describes improvements made to the KyotoEBMT machine translation system. It discusses using forest parsing of input sentences to handle parsing errors and syntactic divergences. It also describes using the Nile alignment tool along with constituent parsing to improve word alignments from the training corpus. New features were added and the reranking was improved by incorporating a neural machine translation-based bilingual language model.

John Richardson - 2015 - KyotoEBMT System Description for the 2nd Workshop on...

Association for Computational Linguistics

El documento describe el sistema de traducción basado en ejemplos KyotoEBMT. El sistema utiliza análisis de dependencia tanto del idioma origen como del idioma destino y puede manejar ambigüedades en las hipótesis de traducción mediante el uso de reglas de rejilla. Los resultados oficiales del WAT2015 muestran mejoras en las métricas BLEU y RIBES con la reranqueación de traducciones, aunque la reranqueación empeora la evaluación humana para la dirección de traducción japonés-chino. El sistema Ky

Zhongyuan Zhu - 2015 - Evaluating Neural Machine Translation in English-Japan...

Association for Computational Linguistics

This document evaluates several neural machine translation models for English to Japanese translation. It finds that simple neural models outperform statistical machine translation baselines. Soft attention models with LSTM units performed best. However, training these models on pre-reordered data hurt performance. The neural models tended to produce grammatically correct but incomplete translations by omitting information. Replacing unknown words helped some models but more sophisticated solutions are needed for models trained on natural order data.

Zhongyuan Zhu - 2015 - Evaluating Neural Machine Translation in English-Japan...

Association for Computational Linguistics

This document evaluates various neural machine translation models for English to Japanese translation. It compares different network architectures, recurrent units, and training data configurations. Results show that soft-attention models outperformed multi-layer encoder-decoder models, and training on pre-reordered data hurt performance. Neural machine translation models tended to generate grammatically correct but incomplete translations.

Hyoung-Gyu Lee - 2015 - NAVER Machine Translation System for WAT 2015

Association for Computational Linguistics

This document describes NAVER's machine translation systems for the WAT 2015 evaluation. For English-to-Japanese translation, the best system combined tree-to-string syntax-based machine translation with neural machine translation re-ranking, achieving a BLEU score of 34.60. For Korean-to-Japanese translation, the top system used phrase-based machine translation and neural machine translation re-ranking, obtaining a BLEU score of 71.38. The document also analyzes the effectiveness of character-level tokenization and other techniques for neural machine translation.

Satoshi Sonoh - 2015 - Toshiba MT System Description for the WAT2015 Workshop

Association for Computational Linguistics

Toshiba presented their machine translation system for the WAT2015 workshop. Their system uses statistical post-editing (SPE) to correct rule-based machine translation (RBMT) output. It also combines SPE and phrase-based statistical machine translation (SMT) results by reranking the merged n-best lists using a recurrent neural network language model. Evaluation showed the combined system achieved the best results on most language pairs compared to SPE and SMT individually. Analysis of system selections by the combination found it primarily chose translations from SPE.

Chenchen Ding - 2015 - NICT at WAT 2015

Association for Computational Linguistics

Graham Neubig - 2015 - Neural Reranking Improves Subjective Quality of Machin...

Association for Computational Linguistics

Neural reranking of machine translation output improves both automatic metrics and subjective human evaluations of translation quality. The document analyzes reranking results from a statistical machine translation system using an attentional neural machine translation model. Reranking corrected errors related to reordering, insertion, deletion, substitution and conjugation. Specifically, it improved phrasal reordering, auxiliary verb insertion/deletion, and coordinate structures. The gains were mainly in grammatical aspects rather than lexical selection. While reranking is shown to be effective, questions remain about comparing it to pure neural machine translation and neural language models.

More from Association for Computational Linguistics (20)