no code implementations • LREC 2016 • Chenhui Chu, Raj Dabre, Sadao Kurohashi
Parallel corpora are crucial for machine translation (MT), however they are quite scarce for most language pairs and domains.
no code implementations • 12 Jan 2017 • Chenhui Chu, Raj Dabre, Sadao Kurohashi
In this paper, we propose a novel domain adaptation method named "mixed fine tuning" for neural machine translation (NMT).
no code implementations • MTSummit 2017 • Raj Dabre, Fabien Cromieres, Sadao Kurohashi
In this paper, we explore a simple solution to "Multi-Source Neural Machine Translation" (MSNMT) which only relies on preprocessing a N-way multilingual corpus without modifying the Neural Machine Translation (NMT) architecture or training procedure.
no code implementations • ACL 2017 • Chenhui Chu, Raj Dabre, Sadao Kurohashi
In this paper, we propose a novel domain adaptation method named {``}mixed fine tuning{''} for neural machine translation (NMT).
1 code implementation • 3 Oct 2017 • Raj Dabre, Sadao Kurohashi
Multilinguality is gradually becoming ubiquitous in the sense that more and more researchers have successfully shown that using additional languages help improve the results in many Natural Language Processing tasks.
1 code implementation • WS 2017 • Fabien Cromieres, Raj Dabre, Toshiaki Nakazawa, Sadao Kurohashi
We describe here our approaches and results on the WAT 2017 shared translation tasks.
no code implementations • IJCNLP 2017 • Fabien Cromieres, Toshiaki Nakazawa, Raj Dabre
Machine Translation (MT) is a sub-field of NLP which has experienced a number of paradigm shifts since its inception.
no code implementations • 14 Jul 2018 • Raj Dabre, Atsushi Fujita
In neural machine translation (NMT), the most common practice is to stack a number of recurrent or feed-forward layers in the encoder and the decoder.
no code implementations • 14 May 2019 • Raj Dabre, Chenhui Chu, Anoop Kunchukuttan
We present a survey on multilingual neural machine translation (MNMT), which has gained a lot of traction in the recent years.
no code implementations • 19 Jun 2019 • Chenhui Chu, Raj Dabre
In this paper, we propose two novel methods for domain adaptation for the attention-only neural machine translation (NMT) model, i. e., the Transformer.
1 code implementation • WS 2019 • Aizhan Imankulova, Raj Dabre, Atsushi Fujita, Kenji Imamura
This paper proposes a novel multilingual multistage fine-tuning approach for low-resource neural machine translation (NMT), taking a challenging Japanese--Russian pair for benchmarking.
no code implementations • WS 2019 • Benjamin Marie, Raj Dabre, Atsushi Fujita
Our primary submission to the task is the result of a simple combination of our SMT and NMT systems.
no code implementations • WS 2019 • Raj Dabre, Kehai Chen, Benjamin Marie, Rui Wang, Atsushi Fujita, Masao Utiyama, Eiichiro Sumita
In this paper, we describe our supervised neural machine translation (NMT) systems that we developed for the news translation task for Kazakh↔English, Gujarati↔English, Chinese↔English, and English→Finnish translation directions.
no code implementations • WS 2019 • Raj Dabre, Eiichiro Sumita
al., 2017) to improve translation quality for Japanese↔English.
no code implementations • 27 Aug 2019 • Raj Dabre, Atsushi Fujita
This paper proposes a novel procedure for training an encoder-decoder based deep neural network which compresses NxM models into a single model enabling us to dynamically choose the number of encoder and decoder layers for decoding.
no code implementations • WS 2019 • Raj Dabre, Eiichiro Sumita
In this paper we describe our submissions to WAT 2019 for the following tasks: English{--}Tamil translation and Russian{--}Japanese translation.
no code implementations • WS 2019 • Toshiaki Nakazawa, Nobushige Doi, Shohei Higashiyama, Chenchen Ding, Raj Dabre, Hideya Mino, Isao Goto, Win Pa Pa, Anoop Kunchukuttan, Yusuke Oda, Shantipriya Parida, Ond{\v{r}}ej Bojar, Sadao Kurohashi
This paper presents the results of the shared tasks from the 6th workshop on Asian translation (WAT2019) including Ja↔En, Ja↔Zh scientific paper translation subtasks, Ja↔En, Ja↔Ko, Ja↔En patent translation subtasks, Hi↔En, My↔En, Km↔En, Ta↔En mixed domain subtasks and Ru↔Ja news commentary translation task.
no code implementations • IJCNLP 2019 • Raj Dabre, Atsushi Fujita, Chenhui Chu
This paper highlights the impressive utility of multi-parallel corpora for transfer learning in a one-to-many low-resource neural machine translation (NMT) setting.
1 code implementation • LREC 2020 • Haiyue Song, Raj Dabre, Atsushi Fujita, Sadao Kurohashi
To address this, we examine a language independent framework for parallel corpus mining which is a quick and effective way to mine a parallel corpus from publicly available lectures at Coursera.
no code implementations • 4 Jan 2020 • Raj Dabre, Chenhui Chu, Anoop Kunchukuttan
We present a survey on multilingual neural machine translation (MNMT), which has gained a lot of traction in the recent years.
no code implementations • 23 Jan 2020 • Haiyue Song, Raj Dabre, Zhuoyuan Mao, Fei Cheng, Sadao Kurohashi, Eiichiro Sumita
To this end, we propose to exploit monolingual corpora of other languages to complement the scarcity of monolingual corpora for the LOI.
no code implementations • WS 2020 • Raj Dabre, Raphael Rubino, Atsushi Fujita
We propose and evaluate a novel procedure for training multiple Transformers with tied parameters which compresses multiple models into one enabling the dynamic choice of the number of encoder and decoder layers during decoding.
1 code implementation • LREC 2020 • Zhuoyuan Mao, Fabien Cromieres, Raj Dabre, Haiyue Song, Sadao Kurohashi
Monolingual pre-training approaches such as MASS (MAsked Sequence to Sequence) are extremely effective in boosting NMT quality for languages with small parallel corpora.
no code implementations • ACL 2020 • Haiyue Song, Raj Dabre, Zhuoyuan Mao, Fei Cheng, Sadao Kurohashi, Eiichiro Sumita
Sequence-to-sequence (S2S) pre-training using large monolingual data is known to improve performance for various S2S NLP tasks.
no code implementations • 20 Sep 2020 • Raj Dabre, Atsushi Fujita
Neural machine translation (NMT) models are typically trained using a softmax cross-entropy loss where the softmax distribution is compared against smoothed gold labels.
no code implementations • COLING 2020 • Abhisek Chakrabarty, Raj Dabre, Chenchen Ding, Masao Utiyama, Eiichiro Sumita
In this study, linguistic knowledge at different levels are incorporated into the neural machine translation (NMT) framework to improve translation quality for language pairs with extremely limited data.
no code implementations • COLING 2020 • Raj Dabre, Chenhui Chu, Anoop Kunchukuttan
The advent of neural machine translation (NMT) has opened up exciting research in building multilingual translation systems i. e. translation models that can handle more than one language pair.
no code implementations • 15 Apr 2021 • Raj Dabre, Aizhan Imankulova, Masahiro Kaneko, Abhisek Chakrabarty
Parallel corpora are indispensable for training neural machine translation (NMT) models, and parallel corpora for most language pairs do not exist or are scarce.
no code implementations • 18 Jun 2021 • Raj Dabre, Atsushi Fujita
Finally, we analyze the effects of recurrently stacked layers by visualizing the attentions of models that use recurrently stacked layers and models that do not.
no code implementations • 25 Aug 2021 • Raj Dabre, Eiichiro Sumita
In this paper we present our open-source neural machine translation (NMT) toolkit called "Yet Another Neural Machine Translation Toolkit" abbreviated as YANMTT which is built on top of the Transformers library.
1 code implementation • Findings (ACL) 2022 • Raj Dabre, Himani Shrotriya, Anoop Kunchukuttan, Ratish Puduppully, Mitesh M. Khapra, Pratyush Kumar
We present IndicBART, a multilingual, sequence-to-sequence pre-trained model focusing on 11 Indic languages and English.
1 code implementation • COLING 2020 • Diptesh Kanojia, Raj Dabre, Shubham Dewangan, Pushpak Bhattacharyya, Gholamreza Haffari, Malhar Kulkarni
We, then, evaluate the impact of our cognate detection mechanism on neural machine translation (NMT), as a downstream task.
Cross-Lingual Information Retrieval Cross-Lingual Word Embeddings +5
no code implementations • 10 Mar 2022 • Aman Kumar, Himani Shrotriya, Prachi Sahu, Raj Dabre, Ratish Puduppully, Anoop Kunchukuttan, Amogh Mishra, Mitesh M. Khapra, Pratyush Kumar
Natural Language Generation (NLG) for non-English languages is hampered by the scarcity of datasets in these languages.
no code implementations • 11 Apr 2022 • Zhengdong Yang, Wangjin Zhou, Chenhui Chu, Sheng Li, Raj Dabre, Raphael Rubino, Yi Zhao
This challenge aims to predict MOS scores of synthetic speech on two tracks, the main track and a more challenging sub-track: out-of-domain (OOD).
no code implementations • Findings (NAACL) 2022 • Zhuoyuan Mao, Chenhui Chu, Raj Dabre, Haiyue Song, Zhen Wan, Sadao Kurohashi
Meanwhile, the contrastive objective can implicitly utilize automatically learned word alignment, which has not been explored in many-to-many NMT.
no code implementations • 6 Jun 2022 • Raj Dabre, Aneerav Sukhoo
In this paper, we describe MorisienMT, a dataset for benchmarking machine translation quality of Mauritian Creole.
1 code implementation • 16 Nov 2022 • Dominik Macháček, Ondřej Bojar, Raj Dabre
There have been several meta-evaluation studies on the correlation between human ratings and offline machine translation (MT) evaluation metrics such as BLEU, chrF2, BertScore and COMET.
1 code implementation • 20 Dec 2022 • Ananya B. Sai, Vignesh Nagarajan, Tanay Dixit, Raj Dabre, Anoop Kunchukuttan, Pratyush Kumar, Mitesh M. Khapra
In this paper, we fill this gap by creating an MQM dataset consisting of 7000 fine-grained annotations, spanning 5 Indian languages and 7 MT systems, and use it to establish correlations between annotator scores and scores obtained using existing automatic metrics.
no code implementations • 19 Apr 2023 • Varun Gumma, Raj Dabre, Pratyush Kumar
Knowledge distillation (KD) is a well-known method for compressing neural models.
2 code implementations • 12 May 2023 • Nandini Mundra, Sumanth Doddapaneni, Raj Dabre, Anoop Kunchukuttan, Ratish Puduppully, Mitesh M. Khapra
However, adapters have not been sufficiently analyzed to understand if PEFT translates to benefits in training/deployment efficiency and maintainability/extensibility.
no code implementations • 16 May 2023 • Zhuoyuan Mao, Raj Dabre, Qianying Liu, Haiyue Song, Chenhui Chu, Sadao Kurohashi
This paper studies the impact of layer normalization (LayerNorm) on zero-shot translation (ZST).
no code implementations • 17 May 2023 • Zhuoyuan Mao, Haiyue Song, Raj Dabre, Chenhui Chu, Sadao Kurohashi
The language-independency of encoded representations within multilingual neural machine translation (MNMT) models is crucial for their generalization ability on zero-shot translation.
1 code implementation • 22 May 2023 • Ratish Puduppully, Anoop Kunchukuttan, Raj Dabre, Ai Ti Aw, Nancy F. Chen
This study investigates machine translation between related languages i. e., languages within the same family that share linguistic characteristics such as word order and lexical similarity.
1 code implementation • 23 May 2023 • Aswanth Kumar, Ratish Puduppully, Raj Dabre, Anoop Kunchukuttan
We learn a regression model, CTQ Scorer (Contextual Translation Quality), that selects examples based on multiple features in order to maximize the translation quality.
2 code implementations • 25 May 2023 • Jay Gala, Pranjal A. Chitale, Raghavan AK, Varun Gumma, Sumanth Doddapaneni, Aswanth Kumar, Janki Nawale, Anupama Sujatha, Ratish Puduppully, Vivek Raghavan, Pratyush Kumar, Mitesh M. Khapra, Raj Dabre, Anoop Kunchukuttan
Prior to this work, there was (i) no parallel training data spanning all 22 languages, (ii) no robust benchmarks covering all these languages and containing content relevant to India, and (iii) no existing translation models which support all the 22 scheduled languages of India.
no code implementations • 26 May 2023 • Dominik Macháček, Peter Polák, Ondřej Bojar, Raj Dabre
Automatic speech translation is sensitive to speech recognition errors, but in a multilingual scenario, the same content may be available in various languages via simultaneous interpreting, dubbing or subtitling.
1 code implementation • 6 Jun 2023 • Zhishen Yang, Raj Dabre, Hideki Tanaka, Naoaki Okazaki
Automating figure caption generation helps move model understandings of scientific documents beyond text and will help authors write informative captions that facilitate communicating scientific findings.
1 code implementation • 27 Jul 2023 • Dominik Macháček, Raj Dabre, Ondřej Bojar
Whisper is one of the recent state-of-the-art multilingual speech recognition and translation models, however, it is not designed for real time transcription.
no code implementations • 31 Jul 2023 • Haiyue Song, Raj Dabre, Chenhui Chu, Sadao Kurohashi, Eiichiro Sumita
Sub-word segmentation is an essential pre-processing step for Neural Machine Translation (NMT).
1 code implementation • 30 Oct 2023 • Heather Lent, Kushal Tatariya, Raj Dabre, Yiyi Chen, Marcell Fekete, Esther Ploeger, Li Zhou, Hans Erik Heje, Diptesh Kanojia, Paul Belony, Marcel Bollmann, Loïc Grobol, Miryam de Lhoneux, Daniel Hershcovich, Michel DeGraff, Anders Søgaard, Johannes Bjerva
Creoles represent an under-explored and marginalized group of languages, with few available resources for NLP research.
1 code implementation • 7 Nov 2023 • Haiyue Song, Raj Dabre, Chenhui Chu, Atsushi Fujita, Sadao Kurohashi
To create the parallel corpora, we propose a dynamic programming based sentence alignment algorithm which leverages the cosine similarity of machine-translated sentences.
no code implementations • 11 Jan 2024 • Aditya Joshi, Raj Dabre, Diptesh Kanojia, Zhuang Li, Haolan Zhan, Gholamreza Haffari, Doris Dippold
Motivated by the performance degradation of NLP models for dialectic datasets and its implications for the equity of language technologies, we survey past research in NLP for dialects in terms of datasets, and approaches.
no code implementations • 13 Jan 2024 • Settaluri Lakshmi Sravanthi, Meet Doshi, Tankala Pavan Kalyan, Rudra Murthy, Pushpak Bhattacharyya, Raj Dabre
To demonstrate this fact, we release a Pragmatics Understanding Benchmark (PUB) dataset consisting of fourteen tasks in four pragmatics phenomena, namely, Implicature, Presupposition, Reference, and Deixis.
no code implementations • 22 Jan 2024 • Pranjal A. Chitale, Jay Gala, Raj Dabre
While we establish the significance of the quality of the target distribution over the source distribution of demonstrations, we further observe that perturbations sometimes act as regularizers, resulting in performance improvements.
no code implementations • 24 Jan 2024 • Wangjin Zhou, Zhengdong Yang, Chenhui Chu, Sheng Li, Raj Dabre, Yi Zhao, Tatsuya Kawahara
We propose MOS-FAD, where MOS can be leveraged at two key points in FAD: training data selection and model fusion.
no code implementations • 25 Jan 2024 • Jaavid Aktar Husain, Raj Dabre, Aswanth Kumar, Jay Gala, Thanmay Jayakumar, Ratish Puduppully, Anoop Kunchukuttan
This study addresses the challenge of extending Large Language Models (LLMs) to non-English languages using non-Roman scripts.
1 code implementation • 26 Jan 2024 • Jay Gala, Thanmay Jayakumar, Jaavid Aktar Husain, Aswanth Kumar M, Mohammed Safi Ur Rahman Khan, Diptesh Kanojia, Ratish Puduppully, Mitesh M. Khapra, Raj Dabre, Rudra Murthy, Anoop Kunchukuttan
We announce the initial release of "Airavata," an instruction-tuned LLM for Hindi.
1 code implementation • 11 Mar 2024 • Mohammed Safi Ur Rahman Khan, Priyam Mehta, Ananth Sankar, Umashankar Kumaravelan, Sumanth Doddapaneni, Suriyaprasaad G, Varun Balan G, Sparsh Jain, Anoop Kunchukuttan, Pratyush Kumar, Raj Dabre, Mitesh M. Khapra
We hope that the datasets, tools, and resources released as a part of this work will not only propel the research and development of Indic LLMs but also establish an open-source blueprint for extending such efforts to other languages.
no code implementations • 20 Mar 2024 • Meet Doshi, Raj Dabre, Pushpak Bhattacharyya
In this paper, we explore the utility of Translationese as synthetic data created using machine translation for pre-training language models (LMs).
no code implementations • 6 Apr 2024 • Poulami Ghosh, Shikhar Vashishth, Raj Dabre, Pushpak Bhattacharyya
How does the importance of positional encoding in pre-trained language models (PLMs) vary across languages with different morphological complexity?
no code implementations • GWC 2016 • Diptesh Kanojia, Raj Dabre, Pushpak Bhattacharyya
India is a country with 22 officially recognized languages and 17 of these have WordNets, a crucial resource.
no code implementations • ACL (WAT) 2021 • Raj Dabre, Abhisek Chakrabarty
The objective of the task was to explore the utility of multilingual approaches using a variety of in-domain and out-of-domain parallel and monolingual corpora.
no code implementations • ACL (WAT) 2021 • Toshiaki Nakazawa, Hideki Nakayama, Chenchen Ding, Raj Dabre, Shohei Higashiyama, Hideya Mino, Isao Goto, Win Pa Pa, Anoop Kunchukuttan, Shantipriya Parida, Ondřej Bojar, Chenhui Chu, Akiko Eriguchi, Kaori Abe, Yusuke Oda, Sadao Kurohashi
This paper presents the results of the shared tasks from the 8th workshop on Asian translation (WAT2021).
no code implementations • WAT 2022 • Raj Dabre
However, to our surprise, we find that existing multilingual NMT systems are able to handle the translation of text annotated with XML tags without any explicit training on data containing said tags.
no code implementations • WMT (EMNLP) 2020 • Raj Dabre, Atsushi Fujita
This paper investigates a combination of SD and TL for training efficient NMT models for ELR settings, where we utilize TL with helping corpora twice: once for distilling the ELR corpora and then during compact model training.
no code implementations • IWSLT 2017 • Raj Dabre, Fabien Cromieres, Sadao Kurohashi
We describe here our Machine Translation (MT) model and the results we obtained for the IWSLT 2017 Multilingual Shared Task.
no code implementations • WAT 2022 • Toshiaki Nakazawa, Hideya Mino, Isao Goto, Raj Dabre, Shohei Higashiyama, Shantipriya Parida, Anoop Kunchukuttan, Makoto Morishita, Ondřej Bojar, Chenhui Chu, Akiko Eriguchi, Kaori Abe, Yusuke Oda, Sadao Kurohashi
This paper presents the results of the shared tasks from the 9th workshop on Asian translation (WAT2022).
no code implementations • COLING 2022 • Abhisek Chakrabarty, Raj Dabre, Chenchen Ding, Hideki Tanaka, Masao Utiyama, Eiichiro Sumita
In this paper we present FeatureBART, a linguistically motivated sequence-to-sequence monolingual pre-training strategy in which syntactic features such as lemma, part-of-speech and dependency labels are incorporated into the span prediction based pre-training framework (BART).
no code implementations • AACL (WAT) 2020 • Raj Dabre, Abhisek Chakrabarty
In this paper we describe our team‘s (NICT-5) Neural Machine Translation (NMT) models whose translations were submitted to shared tasks of the 7th Workshop on Asian Translation.
no code implementations • AACL (WAT) 2020 • Toshiaki Nakazawa, Hideki Nakayama, Chenchen Ding, Raj Dabre, Shohei Higashiyama, Hideya Mino, Isao Goto, Win Pa Pa, Anoop Kunchukuttan, Shantipriya Parida, Ondřej Bojar, Sadao Kurohashi
This paper presents the results of the shared tasks from the 7th workshop on Asian translation (WAT2020).
no code implementations • MTSummit 2021 • Raj Dabre, Aizhan Imankulova, Masahiro Kaneko
To this end and in this paper and we propose wait-k simultaneous document-level NMT where we keep the context encoder as it is and replace the source sentence encoder and target language decoder with their wait-k equivalents.
no code implementations • MTSummit 2021 • Raj Dabre, Atsushi Fujita
In low-resource scenarios and NMT models tend to perform poorly because the model training quickly converges to a point where the softmax distribution computed using logits approaches the gold label distribution.