Atsushi FUJITA, Dr. Eng.

Senior Researcher @
Advanced Translation Technology Lab., ASTREC, NICT, Japan

Research Interests

Research Fields:
Computational Linguistics, Natural Language Processing
Research Topics:
Automatic Paraphrasing, Machine/Human Translation, Others
Interests:
Machine Learning, Data Mining, Social Network Analysis, Text Annotation, Language Communication Support, etc.

Projects and Recent Refereed Publications

MIC-GCP: R&D of Multilingual Speech Translation Technology

  • [Dabre & Fujita, 2019] AAAI. Recurrent Stacking of Layers for Compact Neural Machine Translation Models.
  • [Imamura+, 2018] WNMT. Enhancement of Encoder and Attention Using Target Monolingual Corpora in Neural Machine Translation.
  • [Marie & Fujita, 2018b] AMTA. A Smorgasbord of Features to Combine Phrase-Based and Neural Machine Translation.

Multilingual/Cross-lingual NLP (incl. Machine Translation)

  • [Dabre+, 2020] WNGT. Balancing Cost and Benefit with Tied-Multi Transformers.
  • [Marie+, 2020] ACL. Tagged Back-translation Revisited: Why Does It Really Work?
  • [Marie & Fujita, 2020] ACM TALLIP. Iterative Training of Unsupervised Neural and Statistical Machine Translation Systems.
  • [Song+, 2020] LREC. Coursera Corpus Mining and Multistage Fine-Tuning for Improving Lectures Translation.
  • [Dabre+, 2019] EMNLP-IJCNLP. Exploiting Multilingualism through Multistage Fine-Tuning for Low-Resource Neural Machine Translation.
  • [Imankulova+, 2019] MT Summit. Exploiting Out-of-Domain Parallel Data through Multilingual Transfer Learning for Low-Resource Neural Machine Translation.
  • [Marie & Fujita, 2019b] ACL. Unsupervised Joint Training of Bilingual Word Embeddings.
  • [Marie & Fujita, 2019a] NAACL-HLT. Unsupervised Extraction of Partial Translations for Neural Machine Translation.

Modeling Translation Process

New KAKENHI project led by Prof. Kyo Kageura. 『Developing a Translation Process Model and Constructing an Integrated Translation Environment through Detailed Descriptions of Translation Norms and Competences

  • [Miyata & Fujita, 2017b] JAITS. Investigating the Effectiveness of Pre-Editing Strategy and the Diversity of Pre-Edit Operations for Better Use of Machine Translation (in Japanese).
  • [Fujita+, 2017] LAW. Consistent Classification of Translation Revisions: A Case Study of English-Japanese Student Translations.