no code implementations • ACL ARR May 2021 • Mitchell A Gordon, Kevin Duh, Jared Kaplan
We observe that the development cross-entropy loss of supervised neural machine translation models scales like a power law with the amount of training data and the number of non-embedding parameters in the model.
no code implementations • IWSLT (EMNLP) 2018 • Hirofumi Inaguma, Xuan Zhang, Zhiqi Wang, Adithya Renduchintala, Shinji Watanabe, Kevin Duh
This paper describes the Johns Hopkins University (JHU) and Kyoto University submissions to the Speech Translation evaluation campaign at IWSLT2018.
no code implementations • LREC 2022 • Paul McNamee, Kevin Duh
Translation of the noisy, informal language found in social media has been an understudied problem, with a principal factor being the limited availability of translation corpora in many languages.
no code implementations • IWSLT (ACL) 2022 • Antonios Anastasopoulos, Loïc Barrault, Luisa Bentivogli, Marcely Zanon Boito, Ondřej Bojar, Roldano Cattoni, Anna Currey, Georgiana Dinu, Kevin Duh, Maha Elbayad, Clara Emmanuel, Yannick Estève, Marcello Federico, Christian Federmann, Souhir Gahbiche, Hongyu Gong, Roman Grundkiewicz, Barry Haddow, Benjamin Hsu, Dávid Javorský, Vĕra Kloudová, Surafel Lakew, Xutai Ma, Prashant Mathur, Paul McNamee, Kenton Murray, Maria Nǎdejde, Satoshi Nakamura, Matteo Negri, Jan Niehues, Xing Niu, John Ortega, Juan Pino, Elizabeth Salesky, Jiatong Shi, Matthias Sperber, Sebastian Stüker, Katsuhito Sudoh, Marco Turchi, Yogesh Virkar, Alexander Waibel, Changhan Wang, Shinji Watanabe
The evaluation campaign of the 19th International Conference on Spoken Language Translation featured eight shared tasks: (i) Simultaneous speech translation, (ii) Offline speech translation, (iii) Speech to speech translation, (iv) Low-resource speech translation, (v) Multilingual speech translation, (vi) Dialect speech translation, (vii) Formality control for speech translation, (viii) Isometric speech translation.
no code implementations • IWSLT 2017 • Hao Qin, Takahiro Shinozaki, Kevin Duh
Neural machine translation (NMT) systems have demonstrated promising results in recent years.
no code implementations • EMNLP 2020 • Shuo Sun, Kevin Duh
We present CLIRMatrix, a massively large collection of bilingual and multilingual datasets for Cross-Lingual Information Retrieval extracted automatically from Wikipedia.
no code implementations • MTSummit 2021 • Xuan Zhang, Kevin Duh
A cascaded Sign Language Translation system first maps sign videos to gloss annotations and then translates glosses into a spoken languages.
1 code implementation • EACL (HCINLP) 2021 • Marianna Martindale, Kevin Duh, Marine Carpuat
Successful Machine Translation (MT) deployment requires understanding not only the intrinsic qualities of MT output, such as fluency and adequacy, but also user perceptions.
no code implementations • EMNLP (IWSLT) 2019 • Hirofumi Inaguma, Shun Kiyono, Nelson Enrique Yalta Soplin, Jun Suzuki, Kevin Duh, Shinji Watanabe
In this year, we mainly build our systems based on Transformer architectures in all tasks and focus on the end-to-end speech translation (E2E-ST).
no code implementations • AMTA 2022 • Suzanna Sia, Kevin Duh
We analyze the resulting embeddings’ training dynamics, and where they lie in the embedding space, and show that our trained embeddings can be used for both in-context translation, and diverse generation of the target sentence.
no code implementations • AMTA 2022 • Neha Verma, Kenton Murray, Kevin Duh
Therefore, in this work, we propose two major fine-tuning strategies: our language-first approach first learns the translation language pair via general bitext, followed by the domain via in-domain bitext, and our domain-first approach first learns the domain via multilingual in-domain bitext, followed by the language pair via language pair-specific in-domain bitext.
no code implementations • 7 Mar 2024 • Suzanna Sia, David Mueller, Kevin Duh
Self-supervised large language models have demonstrated the ability to perform Machine Translation (MT) via in-context learning, but little is known about where the model performs the task with respect to prompt instructions and demonstration examples.
no code implementations • 27 Nov 2023 • Elijah Rippeth, Marine Carpuat, Kevin Duh, Matt Post
Lexical ambiguity is a challenging and pervasive problem in machine translation (\mt).
no code implementations • 14 Nov 2023 • Suzanna Sia, Alexandra DeLucia, Kevin Duh
Zero-shot In-context learning is the phenomenon where models can perform the task simply given the instructions.
1 code implementation • 20 Jun 2023 • Cihan Xiao, Henry Li Xinyuan, Jinyi Yang, Dongji Gao, Matthew Wiesner, Kevin Duh, Sanjeev Khudanpur
We introduce HK-LegiCoST, a new three-way parallel corpus of Cantonese-English translations, containing 600+ hours of Cantonese audio, its standard traditional Chinese transcript, and English translation, segmented and aligned at the sentence level.
no code implementations • 12 Jun 2023 • Jeremy Gwinnup, Kevin Duh
Large language models such as BERT and the GPT series started a paradigm shift that calls for building general-purpose models via pre-training on large datasets, followed by fine-tuning on task-specific datasets.
no code implementations • 23 May 2023 • Neha Verma, Kenton Murray, Kevin Duh
Multilingual machine translation has proven immensely useful for low-resource and zero-shot language pairs.
no code implementations • 5 May 2023 • Suzanna Sia, Kevin Duh
In this work which focuses on Machine Translation, we present a perspective of in-context learning as the desired generation task maintaining coherency with its context, i. e., the prompt examples.
no code implementations • 25 Oct 2022 • Kelly Marchisio, Ali Saad-Eldin, Kevin Duh, Carey Priebe, Philipp Koehn
Bilingual lexicons form a critical component of various natural language processing applications, including unsupervised and semisupervised machine translation and crosslingual information retrieval.
1 code implementation • 11 Oct 2022 • Kelly Marchisio, Neha Verma, Kevin Duh, Philipp Koehn
The ability to extract high-quality translation dictionaries from monolingual word embedding spaces depends critically on the geometric similarity of the spaces -- their degree of "isomorphism."
1 code implementation • 20 Jan 2022 • Suraj Nair, Eugene Yang, Dawn Lawrie, Kevin Duh, Paul McNamee, Kenton Murray, James Mayfield, Douglas W. Oard
These models have improved the effectiveness of retrieval systems well beyond that of lexical term matching models such as BM25.
1 code implementation • Findings (EMNLP) 2021 • Kelly Marchisio, Youngser Park, Ali Saad-Eldin, Anton Alyakin, Kevin Duh, Carey Priebe, Philipp Koehn
Alternatively, word embeddings may be understood as nodes in a weighted graph.
no code implementations • 9 Sep 2021 • Hirofumi Inaguma, Yosuke Higuchi, Kevin Duh, Tatsuya Kawahara, Shinji Watanabe
We propose a unified NAR E2E-ST framework called Orthros, which has an NAR decoder and an auxiliary shallow AR decoder on top of the shared encoder.
no code implementations • ACL (IWSLT) 2021 • Hirofumi Inaguma, Brian Yan, Siddharth Dalmia, Pengcheng Guo, Jiatong Shi, Kevin Duh, Shinji Watanabe
This year we made various efforts on training data, architecture, and audio segmentation.
no code implementations • ACL (IWSLT) 2021 • Lei Zhou, Liang Ding, Kevin Duh, Shinji Watanabe, Ryohei Sasano, Koichi Takeda
In the field of machine learning, the well-trained model is assumed to be able to recover the training labels, i. e. the synthetic labels predicted by the model should be as close to the ground-truth labels as possible.
1 code implementation • EACL 2021 • Suzanna Sia, Kevin Duh
Probabilistic topic models in low data resource scenarios are faced with less reliable estimates due to sparsity of discrete word co-occurrence counts, and do not have the luxury of retraining word or topic embeddings using neural methods.
no code implementations • EACL 2021 • Jiatong Shi, Jonathan D. Amith, Rey Castillo Garc{\'\i}a, Esteban Guadalupe Sierra, Kevin Duh, Shinji Watanabe
{``}Transcription bottlenecks{''}, created by a shortage of effective human transcribers (i. e., transcriber shortage), are one of the main challenges to endangered language (EL) documentation.
Automatic Speech Recognition Automatic Speech Recognition (ASR) +1
no code implementations • 26 Jan 2021 • Jiatong Shi, Jonathan D. Amith, Rey Castillo García, Esteban Guadalupe Sierra, Kevin Duh, Shinji Watanabe
"Transcription bottlenecks", created by a shortage of effective human transcribers are one of the main challenges to endangered language (EL) documentation.
Automatic Speech Recognition Automatic Speech Recognition (ASR) +1
no code implementations • 25 Oct 2020 • Hirofumi Inaguma, Yosuke Higuchi, Kevin Duh, Tatsuya Kawahara, Shinji Watanabe
Fast inference speed is an important goal towards real-world deployment of speech translation (ST) systems.
4 code implementations • 18 Aug 2020 • Xiaodong Liu, Kevin Duh, Liyuan Liu, Jianfeng Gao
We explore the application of very deep Transformer models for Neural Machine Translation (NMT).
Ranked #1 on Machine Translation on WMT2014 English-French (using extra training data)
1 code implementation • ACL 2020 • Shuo Sun, Suzanna Sia, Kevin Duh
We present CLIReval, an easy-to-use toolkit for evaluating machine translation (MT) with the proxy task of cross-lingual information retrieval (CLIR).
no code implementations • 8 May 2020 • Shuo Sun, Kevin Duh
Learning to rank is an important task that has been successfully deployed in many real-world information retrieval systems.
no code implementations • LREC 2020 • Kevin Duh, Paul McNamee, Matt Post, Brian Thompson
In this study, we benchmark state of the art statistical and neural machine translation systems on two African languages which do not have large amounts of resources: Somali and Swahili.
1 code implementation • ACL 2020 • Hirofumi Inaguma, Shun Kiyono, Kevin Duh, Shigeki Karita, Nelson Enrique Yalta Soplin, Tomoki Hayashi, Shinji Watanabe
We present ESPnet-ST, which is designed for the quick development of speech-to-speech translation systems in a single framework.
Automatic Speech Recognition Automatic Speech Recognition (ASR) +4
no code implementations • WMT (EMNLP) 2020 • Kelly Marchisio, Kevin Duh, Philipp Koehn
We additionally find that unsupervised MT performance declines when source and target languages use different scripts, and observe very poor performance on authentic low-resource language pairs.
no code implementations • WS 2020 • Mitchell A. Gordon, Kevin Duh
We explore best practices for training small, memory efficient machine translation models with sequence-level knowledge distillation in the domain adaptation setting.
no code implementations • AMTA 2020 • Jason Naradowsky, Xuan Zhang, Kevin Duh
Adapting machine translation systems in the real world is a difficult problem.
1 code implementation • ACL 2020 • Mitchell A. Gordon, Kevin Duh, Nicholas Andrews
Low levels of pruning (30-40%) do not affect pre-training loss or transfer to downstream tasks at all.
no code implementations • TACL 2020 • Xuan Zhang, Kevin Duh
Hyperparameter selection is a crucial part of building neural machine translation (NMT) systems across both academia and industry.
no code implementations • 6 Dec 2019 • Mitchell A. Gordon, Kevin Duh
We then propose an alternative hypothesis under the lens of data augmentation and regularization.
no code implementations • IJCNLP 2019 • Brian Thompson, Rebecca Knowles, Xuan Zhang, Huda Khayrallah, Kevin Duh, Philipp Koehn
Bilingual lexicons are valuable resources used by professional human translators.
1 code implementation • 1 Oct 2019 • Hirofumi Inaguma, Kevin Duh, Tatsuya Kawahara, Shinji Watanabe
In this paper, we propose a simple yet effective framework for multilingual end-to-end speech translation (ST), in which speech utterances in source languages are directly translated to the desired target languages with a universal sequence-to-sequence architecture.
Automatic Speech Recognition Automatic Speech Recognition (ASR) +4
no code implementations • IJCNLP 2019 • Sheng Zhang, Xutai Ma, Kevin Duh, Benjamin Van Durme
We unify different broad-coverage semantic parsing tasks under a transduction paradigm, and propose an attention-based neural framework that incrementally builds a meaning representation via a sequence of semantic relations.
Ranked #2 on UCCA Parsing on SemEval 2019 Task 1
no code implementations • WS 2019 • Matt Post, Kevin Duh
We describe the JHU submissions to the French{--}English, Japanese{--}English, and English{--}Japanese Robustness Task at WMT 2019.
no code implementations • WS 2019 • Tom Lippincott, Pamela Shapiro, Kevin Duh, Paul McNamee
Our submission to the MADAR shared task on Arabic dialect identification employed a language modeling technique called Prediction by Partial Matching, an ensemble of neural architectures, and sources of additional data for training word embeddings and auxiliary language models.
no code implementations • WS 2019 • Pamela Shapiro, Kevin Duh
When translating diglossic languages such as Arabic, situations may arise where we would like to translate a text but do not know which dialect it is.
no code implementations • NAACL 2019 • Brian Thompson, Jeremy Gwinnup, Huda Khayrallah, Kevin Duh, Philipp Koehn
Continued training is an effective method for domain adaptation in neural machine translation.
no code implementations • WS 2019 • Shuoyang Ding, Adithya Renduchintala, Kevin Duh
Most neural machine translation systems are built upon subword units extracted by methods such as Byte-Pair Encoding (BPE) or wordpiece.
1 code implementation • ACL 2019 • Sheng Zhang, Xutai Ma, Kevin Duh, Benjamin Van Durme
Our experimental results outperform all previously reported SMATCH scores, on both AMR 2. 0 (76. 3% F1 on LDC2017T10) and AMR 1. 0 (70. 2% F1 on LDC2014T12).
Ranked #1 on AMR Parsing on LDC2014T12:
no code implementations • NAACL 2019 • Xuan Zhang, Pamela Shapiro, Gaurav Kumar, Paul McNamee, Marine Carpuat, Kevin Duh
We introduce a curriculum learning approach to adapt generic neural machine translation models to a specific domain.
no code implementations • 16 Apr 2019 • Muhammad Mahbubur Rahman, Sorami Hisamoto, Kevin Duh
Community question-answering (CQA) platforms have become very popular forums for asking and answering questions daily.
1 code implementation • TACL 2020 • Sorami Hisamoto, Matt Post, Kevin Duh
Data privacy is an important issue for "machine learning as a service" providers.
1 code implementation • 2 Nov 2018 • Xuan Zhang, Gaurav Kumar, Huda Khayrallah, Kenton Murray, Jeremy Gwinnup, Marianna J. Martindale, Paul McNamee, Kevin Duh, Marine Carpuat
Machine translation systems based on deep neural networks are expensive to train.
no code implementations • 30 Oct 2018 • Sheng Zhang, Xiaodong Liu, Jingjing Liu, Jianfeng Gao, Kevin Duh, Benjamin Van Durme
We present a large-scale dataset, ReCoRD, for machine reading comprehension requiring commonsense reasoning.
Ranked #34 on Common Sense Reasoning on ReCoRD
no code implementations • WS 2018 • Philipp Koehn, Kevin Duh, Brian Thompson
We report on the efforts of the Johns Hopkins University to develop neural machine translation systems for the shared task for news translation organized around the Conference for Machine Translation (WMT) 2018.
no code implementations • EMNLP 2018 • Sheng Zhang, Xutai Ma, Rachel Rudinger, Kevin Duh, Benjamin Van Durme
We introduce the task of cross-lingual decompositional semantic parsing: mapping content provided in a source language into a decompositional semantic analysis based on a target language.
5 code implementations • 24 Sep 2018 • Xiaodong Liu, Wei Li, Yuwei Fang, Aerin Kim, Kevin Duh, Jianfeng Gao
This paper presents an extension of the Stochastic Answer Network (SAN), one of the state-of-the-art machine reading comprehension models, to be able to judge whether a question is unanswerable or not.
1 code implementation • WS 2018 • Brian Thompson, Huda Khayrallah, Antonios Anastasopoulos, Arya D. McCarthy, Kevin Duh, Rebecca Marvin, Paul McNamee, Jeremy Gwinnup, Tim Anderson, Philipp Koehn
To better understand the effectiveness of continued training, we analyze the major components of a neural machine translation system (the encoder, decoder, and each embedding space) and consider each component's contribution to, and capacity for, domain adaptation.
no code implementations • WS 2019 • Adithya Renduchintala, Pamela Shapiro, Kevin Duh, Philipp Koehn
Neural machine translation (NMT) systems operate primarily on words (or sub-words), ignoring lower-level patterns of morphology.
no code implementations • 5 Sep 2018 • Pamela Shapiro, Kevin Duh
Neural Machine Translation (NMT) in low-resource settings and of morphologically rich languages is made difficult in part by data sparsity of vocabulary words.
1 code implementation • WS 2018 • Huda Khayrallah, Brian Thompson, Kevin Duh, Philipp Koehn
Supervised domain adaptation{---}where a large generic corpus and a smaller in-domain corpus are both available for training{---}is a challenge for neural machine translation (NMT).
no code implementations • 5 Jun 2018 • Shuoyang Ding, Kevin Duh
Using pre-trained word embeddings as input layer is a common practice in many natural language processing (NLP) tasks, but it is largely neglected for neural machine translation (NMT).
no code implementations • NAACL 2018 • Shota Sasaki, Shuo Sun, Shigehiko Schamoni, Kevin Duh, Kentaro Inui
Cross-lingual information retrieval (CLIR) is a document retrieval task where the documents are written in a language different from that of the user{'}s query.
no code implementations • WS 2018 • Pamela Shapiro, Kevin Duh
Neural machine translation has achieved impressive results in the last few years, but its success has been limited to settings with large amounts of parallel data.
no code implementations • SEMEVAL 2018 • Hongyuan Mei, Sheng Zhang, Kevin Duh, Benjamin Van Durme
Cross-lingual information extraction (CLIE) is an important and challenging task, especially in low resource scenarios.
3 code implementations • 21 Apr 2018 • Xiaodong Liu, Kevin Duh, Jianfeng Gao
We propose a stochastic answer network (SAN) to explore multi-step inference strategies in Natural Language Inference.
Ranked #32 on Natural Language Inference on SNLI
no code implementations • 21 Apr 2018 • Sheng Zhang, Kevin Duh, Benjamin Van Durme
We introduce the task of cross-lingual semantic parsing: mapping content provided in a source language into a meaning representation based on a target language.
1 code implementation • SEMEVAL 2018 • Sheng Zhang, Kevin Duh, Benjamin Van Durme
Fine-grained entity typing is the task of assigning fine-grained semantic types to entity mentions.
5 code implementations • ACL 2018 • Xiaodong Liu, Yelong Shen, Kevin Duh, Jianfeng Gao
We propose a simple yet robust stochastic answer network (SAN) that simulates multi-step reasoning in machine reading comprehension.
Ranked #24 on Question Answering on SQuAD1.1 dev
no code implementations • IJCNLP 2017 • Yelong Shen, Xiaodong Liu, Kevin Duh, Jianfeng Gao
Using a state-of-the-art RC model, we empirically investigate the performance of single-turn and multiple-turn reasoning on the SQuAD and MS MARCO datasets.
no code implementations • IJCNLP 2017 • Benjamin Van Durme, Tom Lippincott, Kevin Duh, Deana Burchfield, Adam Poliak, Cash Costello, Tim Finin, Scott Miller, James Mayfield, Philipp Koehn, Craig Harman, Dawn Lawrie, Ch May, ler, Max Thomas, Annabelle Carrell, Julianne Chaloux, Tongfei Chen, Alex Comerford, Mark Dredze, Benjamin Glass, Shudong Hao, Patrick Martin, Pushpendre Rastogi, Rashmi Sankepally, Travis Wolfe, Ying-Ying Tran, Ted Zhang
It combines a multitude of analytics together with a flexible environment for customizing the workflow for different users.
no code implementations • IJCNLP 2017 • Sheng Zhang, Kevin Duh, Benjamin Van Durme
Cross-lingual open information extraction is the task of distilling facts from the source language into representations in the target language.
no code implementations • IJCNLP 2017 • Ryan Cotterell, Kevin Duh
Low-resource named entity recognition is still an open problem in NLP.
Cross-Lingual Transfer Low Resource Named Entity Recognition +4
no code implementations • IJCNLP 2017 • Aaron Steven White, Pushpendre Rastogi, Kevin Duh, Benjamin Van Durme
We propose to unify a variety of existing semantic classification tasks, such as semantic role labeling, anaphora resolution, and paraphrase detection, under the heading of Recognizing Textual Entailment (RTE).
no code implementations • IJCNLP 2017 • Huda Khayrallah, Gaurav Kumar, Kevin Duh, Matt Post, Philipp Koehn
Domain adaptation is a major challenge for neural machine translation (NMT).
no code implementations • IJCNLP 2017 • Dingquan Wang, Nanyun Peng, Kevin Duh
We show how to adapt bilingual word embeddings (BWE{'}s) to bootstrap a cross-lingual name-entity recognition (NER) system in a language with no labeled data.
2 code implementations • 24 Apr 2017 • Chandler May, Kevin Duh, Benjamin Van Durme, Ashwin Lall
We develop a streaming (one-pass, bounded-memory) word embedding algorithm based on the canonical skip-gram with negative sampling algorithm implemented in word2vec.
no code implementations • EACL 2017 • Sheng Zhang, Kevin Duh, Benjamin Van Durme
Conventional pipeline solutions decompose the task as machine translation followed by information extraction (or vice versa).
4 code implementations • 15 Jan 2017 • Graham Neubig, Chris Dyer, Yoav Goldberg, Austin Matthews, Waleed Ammar, Antonios Anastasopoulos, Miguel Ballesteros, David Chiang, Daniel Clothiaux, Trevor Cohn, Kevin Duh, Manaal Faruqui, Cynthia Gan, Dan Garrette, Yangfeng Ji, Lingpeng Kong, Adhiguna Kuncoro, Gaurav Kumar, Chaitanya Malaviya, Paul Michel, Yusuke Oda, Matthew Richardson, Naomi Saphra, Swabha Swayamdipta, Pengcheng Yin
In the static declaration strategy that is used in toolkits like Theano, CNTK, and TensorFlow, the user first defines a computation graph (a symbolic representation of the computation), and then examples are fed into an engine that executes this computation and computes its derivatives.
no code implementations • TACL 2017 • Sheng Zhang, Rachel Rudinger, Kevin Duh, Benjamin Van Durme
Humans have the capacity to draw common-sense inferences from natural language: various things that are likely but not certain to hold based on established discourse, and are rarely stated explicitly.
1 code implementation • 7 Aug 2016 • Keisuke Sakaguchi, Kevin Duh, Matt Post, Benjamin Van Durme
Inspired by the findings from the Cmabrigde Uinervtisy effect, we propose a word recognition model based on a semi-character level recurrent neural network (scRNN).
no code implementations • 22 Apr 2016 • Adhiguna Kuncoro, Yuichiro Sawai, Kevin Duh, Yuji Matsumoto
We propose a transition-based dependency parser using Recurrent Neural Networks with Long Short-Term Memory (LSTM) units.
no code implementations • 16 Aug 2015 • Kaisheng Yao, Trevor Cohn, Katerina Vylomova, Kevin Duh, Chris Dyer
This gate is a function of the lower layer memory cell, the input to and the past memory cell of this layer.
no code implementations • 18 Dec 2014 • Daniel Fried, Kevin Duh
We investigate the hypothesis that word representations ought to incorporate both distributional and relational semantics.
no code implementations • 14 Dec 2014 • Daniel Fried, Kevin Duh
We investigate the hypothesis that word representations ought to incorporate both distributional and relational semantics.
no code implementations • LREC 2014 • Fei Cheng, Kevin Duh, Yuji Matsumoto
Synthetic word analysis is a potentially important but relatively unexplored problem in Chinese natural language processing.