SMT

Neural Machine Translation for Low Resource Languages using Bilingual Lexicon Induced from Comparable Corpora (N18-4)

Challenge:	This paper describes the approach developed by the Tilburg University team for the shallow track of the Multilingual Surface Realization Shared Task 2019 (SR'19).
Approach:	They propose a method for the shallow track of the Multilingual Surface Realization Shared Task 2019 using a rule-based and a statistical machine translation (SMT) model.
Outcome:	The proposed approach can generate texts in 11 languages, compared with the submission of the same approach for the same task in 2018 which only covered 6 languages.

Copied to clipboard

Sree Harsha Ramesh, Krishna Prasad Sankaranarayanan

An Effective Approach to Unsupervised Machine Translation (P19-1)

Challenge:	Recent crowdsourcing efforts and workshops on machine translation have resulted in small amounts of parallel texts for building viable machine translation systems for low resource pairs.
Approach:	They propose to use an end-to-end Siamese bidirectional recurrent neural network to extract parallel sentences from Wikipedia to improve BLEU scores on both NMT and phrase-based SMT systems.
Outcome:	The proposed approach improves BLEU scores on both NMT and phrase-based SMT systems for the low-resource language pairs English–Hindi and English–Tamil when compared to training exclusively on the limited bilingual corpora.

Copied to clipboard

Mikel Artetxe, Gorka Labaka, Eneko Agirre

Rethinking Round-Trip Translation for Machine Translation Evaluation (2023.findings-acl)

Challenge:	a recent research line has managed to train both unsupervised and unsupervised machine translation systems using monolingual corpora only.
Approach:	They propose to use monolingual corpora to train both unsupervised and unsupervised machine translation systems.
Outcome:	The proposed system achieves 22.5 BLEU points in English-to-German WMT 2014, 5.5 points more than the previous best unsupervised system, and 0.5 points more in the (supervised) shared task winner back in 2014.

Copied to clipboard

Terry Yue Zhuo, Qiongkai Xu, Xuanli He, Trevor Cohn

Construction of an Evaluation Corpus for Grammatical Error Correction for Learners of Japanese as a Second Language (2020.lrec-1)

Challenge:	Automatic evaluation methods for translation often require model training and the availability of parallel corpora limits their applicability to low-resource settings.
Approach:	They revisit the statistical machine translation technique and use it to improve translation quality.
Outcome:	The proposed method improves translation quality estimation models and identifies adversarial competitors in shared tasks via cross-system verification.

Copied to clipboard

Aomi Koyama, Tomoshige Kiyuna, Kenji Kobayashi, Mio Arai, Mamoru Komachi

ChrEnTranslate: Cherokee-English Machine Translation Demo with Quality Estimation and Corrective Feedback (2021.acl-demo)

Challenge:	The Lang-8 corpus is suitable as a training dataset for machine translation-based grammatical error correction systems but it is not suitable as an evaluation dataset because corrected sentences sometimes include inappropriate sentences.
Approach:	They created an evaluation corpus for correcting grammatical errors made by Japanese as a second language learners using neural machine translation and statistical machine translation techniques.
Outcome:	The proposed corpus has less noise and its annotation scheme reflects the characteristics of the dataset, making it ideal for correcting grammatical errors in sentences written by learners of Japanese as a Second Language (JSL).

Copied to clipboard

Shiyue Zhang, Benjamin Frey, Mohit Bansal

Learning Source Phrase Representations for Neural Machine Translation (2020.acl-main)

Challenge:	a new study shows that machine translation models can translate fragments of the source sentence but make major mistakes.
Approach:	They propose an online machine translation demonstration system for translation between English and an endangered language Cherokee.
Outcome:	The proposed system achieves state-of-the-art translation performance and improves quality estimation . the proposed system can translate between English and an endangered language Cherokee .

Copied to clipboard

Hongfei Xu, Josef van Genabith, Deyi Xiong, Qiuhui Liu, Jingyi Zhang

Semantic Text Processing

Challenge:	Existing approaches to machine translation have been shown to be effective for long sentences . however, the attentional network can't capture long-distance dependencies .
Approach:	They propose a multi-head attention mechanism which generates phrase representations from token representations and incorporates them into the Transformer translation model to enhance its ability to capture long-distance relationships.
Outcome:	The proposed model can be computed in parallel and improves on the WMT 14 tasks.

ChrEn: Cherokee-English Machine Translation for Endangered Language Revitalization (2020.emnlp-main)

Copied to clipboard

Shiyue Zhang, Benjamin Frey, Mohit Bansal

Near Human-Level Performance in Grammatical Error Correction with Hybrid Machine Translation (N18-2)

Challenge:	Cherokee is a highly endangered Native American language spoken by the Cherokee people . there are only 2,000 fluent first language Cherokee speakers remaining in the world .
Approach:	They propose a Cherokee-English parallel dataset to facilitate machine translation between Cherokee and English.
Outcome:	The proposed dataset compares Cherokee-English and English-Cherokee machine translation systems . the results show that the datasets are low-resource and low-cost compared to other datasets .

Copied to clipboard

Roman Grundkiewicz, Marcin Junczys-Dowmunt

Modeling Coherence for Neural Machine Translation with Dynamic and Topic Caches (C18-1)

Challenge:	Currently, most effective GEC systems are based on phrase-based statistical machine translation.
Approach:	They combine two of the most popular approaches to automated Grammatical Error Correction (GEC) they create a hybrid GEC system that preserves the accuracy of SMT output and generates more fluent sentences .
Outcome:	The proposed system achieves state-of-the-art on the CoNLL-2014 and JFLEG benchmarks.

Copied to clipboard

Shaohui Kuang, Deyi Xiong, Weihua Luo, Guodong Zhou

Approaching Neural Grammatical Error Correction as a Low-Resource Machine Translation Task (N18-1)

Challenge:	Current neural machine translation systems translate a text sentence-by-sentence, ignoring cross-sentent links and dependencies.
Approach:	They propose a cache-based approach to modeling coherence for neural machine translation . they capture contextual information either from recently translated sentences or the entire document .
Outcome:	The proposed model improves on state-of-the-art translation models on many languages . it captures contextual information from recently translated sentences or the entire document .

Copied to clipboard

Marcin Junczys-Dowmunt, Roman Grundkiewicz, Shubha Guha, Kenneth Heafield

Can Cognate Prediction Be Modelled as a Low-Resource Machine Translation Task? (2021.findings-acl)

Challenge:	Previously, neural methods in grammatical error correction did not reach state-of-the-art results compared to phrase-based statistical machine translation (SMT) systems that improve on results by SMT use their set-up as a backbone for more complex systems.
Approach:	They propose a set of model-independent methods for neural GEC that can be easily applied in most GEC settings.
Outcome:	The proposed methods outperform state-of-the-art neural GEC systems by 10% M2 on the CoNLL-2014 benchmark and 5.9% on the JFLEG test set.

Copied to clipboard

Clémentine Fourrier, Rachel Bawden, Benoît Sagot

Forest-Based Neural Machine Translation (P18-1)

Challenge:	Existing work on cognate prediction based on similarities of two languages has not studied their differences or optimized architectural choices.
Approach:	They compare statistical and neural MT architectures to a bilingual setup to test their hypothesis . they use monolingual pretraining, backtranslation and multilinguality to test the hypothesis based on the results .
Outcome:	The proposed architectures can be used to generate cognates in a given language . the proposed architecture can be employed with monolingual pretraining, backtranslation and multilinguality .

Copied to clipboard

Chunpeng Ma, Akihiro Tamura, Masao Utiyama, Tiejun Zhao, Eiichiro Sumita

English-Basque Statistical and Neural Machine Translation (L18-1)

Challenge:	Compared with string-to-string systems, tree-based NMT methods use more syntactic information and can incorporate prior knowledge.
Approach:	They propose a tree-based neural machine translation method that translates a linearized packed forest under a simple sequence-to-sequence framework.
Outcome:	The proposed method outperforms tree-based approaches in the BLEU score of the proposed model.

Copied to clipboard

Inigo Jauregi Unanue, Lierni Garmendia Arratibel, Ehsan Zare Borzeshi, Massimo Piccardi

Bilingual Dictionary Based Neural Machine Translation without Using Parallel Sentences (2020.acl-main)

Challenge:	Neural machine translation (NMT) requires large training corpora, which is problematic for low-resource languages.
Approach:	They propose to use an open-domain and an IT-domain corpora to train machine translations in English-Basque.
Outcome:	The proposed systems outperform OpenNMT, Moses SMT and Google Translate in English-Basque translation.

Copied to clipboard

Xiangyu Duan, Baijun Ji, Hao Jia, Min Tan, Min Zhang, Boxing Chen, Weihua Luo, Yue Zhang

Translating Web Search Queries into Natural Language Questions (L18-1)

Challenge:	a monolingual speaker can learn to translate by looking up a bilingual dictionary . a novel task of machine translation (MT) is based on no parallel sentences but can refer to a ground-truth bilingual dictionary and large-scale monolingual corpora.
Approach:	They propose a task of machine translation that uses a bilingual dictionary and large-scale monolingual corpora to translate a monolingual speaker.
Outcome:	The proposed task is based on a bilingual dictionary and large scale monolingual corpora, while being independent on parallel sentences.

Copied to clipboard

Adarsh Kumar, Sandipan Dandapat, Sushil Chordia

Pipeline Signed Japanese Translation Focusing on a Post-positional Particle Complement and Conjugation in a Low-resource Setting (2021.findings-acl)

Challenge:	a new method to generate natural language questions from keyword-based queries is proposed . a synergy between query-to-question problem and standard machine translation (MT) model is found .
Approach:	They propose a method to generate well-formed natural language questions from keyword-based queries.
Outcome:	The proposed method is well-formed natural language question generated from keyword-based query.

Copied to clipboard

Ken Yano, Akira Utsumi

Can Automatic Post-Editing Improve NMT? (2020.emnlp-main)

Challenge:	a pipeline translation method is proposed to take advantage of the similarities and differences between sign language and spoken language.
Approach:	They propose a pipeline translation method that takes advantage of similarities between spoken and spoken Japanese . they map glosses to spoken language words and train them using a monolingual Japanese corpus .
Outcome:	The proposed method performs robustly on the low-resource corpus and is +4.4/+4.9 points above baseline.

Copied to clipboard

Shamil Chollampatt, Raymond Hendy Susanto, Liling Tan, Ewa Szymanska

Parallel Corpora for bi-lingual English-Ethiopian Languages Statistical Machine Translation (C18-1)

Challenge:	APE has been successful with statistical machine translation systems but has not been as successful over neural machine translation (NMT) systems.
Approach:	They propose to train neural APE models on a corpus of human post-edits of NMT and compile a larger corpus to test their hypothesis.
Outcome:	The proposed model can improve a strong in-domain NMT system, challenging the current understanding in the field.

Copied to clipboard

Solomon Teferra Abate, Michael Melese, Martha Yifiru Tachbelie, Million Meshesha, Solomon Atinafu, Wondwossen Mulugeta, Yaregal Assabie, Hafte Abera, Binyam Ephrem, Tewodros Abebe, Wondimagegnhue Tsegaye, Amanuel Lemma, Tsegaye Andargie, Seifedin Shifaw

Anticipating Future with Large Language Model for Simultaneous Machine Translation (2025.naacl-long)

Challenge:	Various approaches to machine translation have been and are being used in the research community, that can broadly classified as rule-based and corpus based.
Approach:	They propose to develop parallel corpora for English and Ethiopian languages such as Amharic, Tigrigna, Afan-Oromo, Wolaytta and Ge’ez.
Outcome:	The proposed system improves on the English-Ethiopian languages.

Copied to clipboard

Siqi Ouyang, Oleksii Hrinchuk, Zhehuai Chen, Vitaly Lavrukhin, Jagadeesh Balam, Lei Li, Boris Ginsburg

The Nunavut Hansard Inuktitut–English Parallel Corpus 3.0 with Preliminary Machine Translation Results (2020.lrec-1)

Challenge:	Existing methods only use the partial utterance that has already arrived at the input and the generated hypothesis.
Approach:	They propose to use a large language model to predict future source words and opportunistically translate without introducing too much risk.
Outcome:	The proposed method outperforms baselines on four language directions and achieves the best translation quality-latency trade-off by up to 5 BLEU points at the same latency.

Copied to clipboard

Eric Joanis, Rebecca Knowles, Roland Kuhn, Samuel Larkin, Patrick Littell, Chi-kiu Lo, Darlene Stewart, Jeffrey Micher

A Retrieve-and-Rewrite Initialization Method for Unsupervised Machine Translation (2020.acl-main)

Challenge:	Inuktitut language is a member of the Inuit-Yupik-Unangan family . it is spoken in two territories, Nunavut and the Northwest Territories .
Approach:	They describe a sentence-aligned Inuktitut–English corpus released in Nunavut . it is the largest parallel corpus of a polysynthetic language released to date . they also describe preliminary experiments on machine translation between the languages .
Outcome:	The proposed corpus is the largest sentence-aligned corpus of a polysynthetic language or an Indigenous language of the Americas . the alignments were evaluated and the results were compared with other methods .

Copied to clipboard

Shuo Ren, Yu Wu, Shujie Liu, Ming Zhou, Shuai Ma

Benchmarking Neural and Statistical Machine Translation on Low-Resource African Languages (2020.lrec-1)

Challenge:	Recent work shows successful methods for unsupervised machine translation (UMT) initialization stage is important since bad initialization may wrongly squeeze the search space and too much noise may hurt the final performance.
Approach:	They propose a retrieval and rewriting based method to better initialize unsupervised translation models.
Outcome:	The proposed method improves translation performance by over 4 BLEU scores.

Copied to clipboard

Kevin Duh, Paul McNamee, Matt Post, Brian Thompson

Unsupervised Statistical Machine Translation (D18-1)

Challenge:	a recent study has focused on languages where large amounts of resources are available.
Approach:	They benchmark state of the art statistical and neural machine translation systems on Somali and Swahili languages . they find that statistical machine translation and neural translation can perform similarly in low-resource scenarios .
Outcome:	The results show that statistical machine translation and neural machine translation perform similarly in low-resource scenarios.

Copied to clipboard

Mikel Artetxe, Gorka Labaka, Eneko Agirre

Neural Grammatical Error Correction with Finite State Transducers (N19-1)

Challenge:	Neural Machine Translation (NMT) systems can be trained from monolingual corpora without supervision.
Approach:	They propose a phrase-based approach that trains from monolingual corpora . their method is based on phrase-driven Statistical Machine Translation (SMT) they propose to train NMT systems without supervision from monolinguistic corpors .
Outcome:	The proposed approach improves on the existing supervised systems by combining a phrase table with an n-gram language model and fine-tuning hyperparameters through an unsupervised MERT variant.

Copied to clipboard

Felix Stahlberg, Christopher Bryant, Bill Byrne

Syntactic Text Processing

Challenge:	Language model based GEC (LM-GEC) is a promising alternative to SMT and neural sequence-to-sequence models.
Approach:	They propose to use finite state transducers to improve LM-GEC by rescoring with neural language models.
Outcome:	The proposed model outperforms the best published results on the CoNLL-2014 test set and achieves far better relative improvements over the baselines.

Morphology Injection for English-Malayalam Statistical Machine Translation (L18-1)

Copied to clipboard

Sreelekha S, Pushpak Bhattacharyya

Linguistically-Motivated Yorùbá-English Machine Translation (2022.coling-1)

Challenge:	Statistical Machine Translation fails to handle the rich morphology when translating into morphologically rich language.
Approach:	They propose a method to generate unseen morphological forms from the parallel corpus . they propose morphology injection method to enrich the corpus with generated morphologies .
Outcome:	The proposed method improves the quality of the translation in English-Malayalam.

Copied to clipboard

Ife Adebara, Muhammad Abdul-Mageed, Miikka Silfverberg

Extracting an English-Persian Parallel Corpus from Comparable Corpora (L18-1)

Challenge:	Several phenomena where asymmetry arises have been identified as challenging problems for machine translation.
Approach:	They perform a fine-grained analysis of how an SMT system compares with two NMT systems when translating bare nouns into English.
Outcome:	The proposed model outperforms the SMT and BiLSTM models for 4 categories and the BiLST outperformed the SLT models for 3 categories.

Copied to clipboard

Akbar Karimi, Ebrahim Ansari, Bahram Sadeghi Bigham

Improving domain-specific SMT for low-resourced languages using data from different domains (L18-1)

Challenge:	Existing methods to extract parallel sentences from Wikipedia are limited for some language pairs such as Persian-English.
Approach:	They propose a bidirectional method to extract parallel sentences from Wikipedia . they add extracted sentences to existing training data and use IR system to measure similarity .
Outcome:	The proposed method outperforms the one-directional approach in analyzing translation data from two translation systems and IR systems.

Copied to clipboard

Fathima Farhath, Pranavan Theivendiram, Surangika Ranathunga, Sanath Jayasena, Gihan Dias

Document Translation vs. Query Translation for Cross-Lingual Information Retrieval in the Medical Domain (2020.acl-main)

Challenge:	Evaluation of domain-specific statistical machine translation system for official government letters . use of pseudo in-domain data showed improvement for both test sets .
Approach:	They develop a statistical machine translation system for official government letters . the system is based on a parallel in-domain dataset containing official letters based in Sinhala and Tamil .
Outcome:	The proposed system improves on the in-domain data in the domain of official government letters . the evaluations show that the system requires quality data from diverse subject matters and sources to perform better.

Copied to clipboard

Shadi Saleh, Pavel Pecina

Language Modeling, Lexical Translation, Reordering: The Training Process of NMT through the Lens of Classical SMT (2021.emnlp-main)

Challenge:	Existing studies of document translation and query translation are outdated and do not reflect the current advances in machine translation.
Approach:	They compare document translation and query translation approaches to cross-lingual information retrieval . they exploit Statistical Machine Translation and Neural Machine Translation paradigms to translate queries into English and English .
Outcome:	The proposed approach outperforms the DT approach in translation quality and retrieval quality.

Copied to clipboard

Elena Voita, Rico Sennrich, Ivan Titov