ニューラル機械翻訳は，大量の対訳データを教師データとしてニューラルネットに基づくモデルを学習し，機械翻訳を実現する技術です．大量の対訳データが手に入る言語対については，人間の翻訳者に迫る翻訳精度を達成しつつありますが，対訳の手に入りにくい低資源言語の翻訳は難しいのが現状です．また，SNSのくだけた表現ように対訳データがほとんど存在しない分野の翻訳は苦手でもあります．そこで対訳になっていない単言語データ（教師なしデータ）を対訳データ（教師データ）と併用してニューラル翻訳モデルを学習する半教師あり学習手法の研究をしています．また，近年では単言語データで事前学習した大規模言語モデル (LLM: Large Language Model) も利用できるようになってきました．これを併用する翻訳手法も研究しています．
　開発した技術は，低言語資源の翻訳，手話の翻訳，特徴量を言語化するタスク（例：健康診断結果の自動所見作成）などに応用していきます．

スマートデザイン概論，オープンイノベーション論，人間中心デザインⅠ，人間機械コミュニケーションプロトタイピングⅠ，人間機械コミュニケーションプロトタイピングⅡ，情報処理基礎Ⅱ，ICT研究専門ゼミナールⅠ，ICT研究専門ゼミナールⅡ，卒業研究Ⅰ，卒業研究Ⅱ，デザイン学ゼミナール（大学院）

2010年〜2012年, アジア太平洋機械翻訳協会(AAMT), 機械翻訳課題調査委員会委員
2012年〜2014年, 人工知能学会, 理事
　・2013年, 全国大会実行委員会, 副委員長
　・2014年, 全国大会実行委員会, 委員長
2013年〜2014年, 日本特許情報機構, 調査委員会委員
2014年〜2017年, 言語処理学会, 代議員
2015年〜2017年, 人工知能学会, 代議員
論文査読（言語処理学会, 電子情報通信学会, 人工知能学会など）
プログラム委員
　・ACL 2008, 2013, 2014
　・EMNLP 2010
　・IWSLT 2009, 2012, 2014, 2017
　・MT Summit 2011, 2023
　・AMTA 2024
　・ICAICTA 2020, 2025など

[1] 森田 知熙, 秋葉 友良, 塚田 元, “ニューラル機械翻訳の反復的逆翻訳に基づくデータ拡張のための混成サンプリング手法,” 電子情報通信学会論文誌D-II, Vol.J83-D-II, No.11, pp.2457-2464 (2023).
[2] 平尾 努, 磯崎 秀樹, 須藤 克仁, Duh Kevin, 塚田 元, 永田 昌明, “語順の相関に基づく機械翻訳の自動評価法,” 自然言語処理, Vol.21, No.3, pp.421-444 (2014).
[3] 林 克彦, 須藤 克仁, 塚田 元, 鈴木 潤, 永田 昌明, “単語並べ替えと冠詞生成の同時逐次処理：日英機械翻訳への適用,” 自然言語処理, Vol.21, No.5, pp.1037-1057 (2014).
[4] Katsuhito Sudoh, Xianchao Wu, Kevin Duh, and Hajime Tsukada, “Syntax-based post-ordering. for efficient Japanese-to-English translation,” ACM Transactions on Asian Language Information Processing (TALIP), Vol.12, No.3, pp.1-15 (2013).
[5] Hideki Isozaki, Katsuhito Sudoh, Hajime Tsukada, and Kevin Duh, “HPSG-based preprocessing. for English-to-Japanese translation,” ACM Transactions on Asian Language Information Processing (TALIP), Vol.11, Vol.3, pp.1-16 (2012).
[6] 林 克彦, 渡辺 太郎, 塚田 元, 磯崎 秀樹, “統計的機械翻訳のためのマージン最大化学習,” 人工知能学会論文誌, Vol.25, No.5, pp.593-601 (2010).
[7] Katsuhito Sudoh, Hajime Tsukada, and Hideki Isozaki, “Named entity recognition from speech. using discriminative models and speech recognition confidence,” Journal of Information Processing, Vol.17, pp.72-81 (2009).
[8] Atsushi Nakamura, Masaki Naito, Hajime Tsukada, Rainer Gruhn, Eiichiro Sumita, Hideki. Kashioka, Hideharu Nakajima, Tohru Shimizu, and Yoshinori Sagisaka, “A speech translation system applied to a real-world task/domain and its evaluation using real-world speech data,” IEICE Transactions on Information and Systems, Vol.E84-D, No.1, pp.142-154 (2001).
[9] 塚田　元, “重み付き有限状態トランスデューサの自動学習と発音変形のモデル化,” 電子情報通信学会論文誌, Vol.J83-D-II, No.11, pp.2457-2464 (2000).
[10] Hajime Tsukada, Hirofumi Yamamoto, Toshiyuki Takezawa, and Yoshinori Sagisaka, “Reliable. utterance segment recognition by integrating a grammar with statistical language constraints,” Speech Communication, Vol.26, No.4, pp.299-309 (1998).
[11] Mikio Kitai, Kazuo Hakoda, Shigeki Sagayama, Tomokazu Yamada, Hajime Tsukada, Satoshi Takahashi, Yoshiaki Noda, Jun ichi Takahashi, Yuki Yoshida, Kazuhiro Arai, Takashi Imoto, and Tomohisa Hirokawa, “ASR and TTS telecommunications applications in Japan,” Speech Communication, Vol.23, No.1-2, pp.17-30 (1997).

[12] Shoki Hamada, Tomoyoshi Akiba, and Hajime Tsukada, “AkibaNLP-TUT: Injecting Language-Specific Word-Level Noise for Low-Resource Language Translation,” 10th Conference on Machine Translation (WMT 2025), (2025).
[13] Taisei Sone, Tomoyoshi Akiba and Hajime Tsukada, “Incorporating Curriculum Learning into Iterative Back-Translation for Neural Machine Translation,” 12th International Conference on Advanced Informatics: Concepts, Theory and Applications (ICAICTA 2025), (2025).
[14] Takuma Tanigawa, Tomoyosi Akiba, and Hajime Tsukada, “Analysis on Unsupervised. Acquisition Process of Bilingual Vocabulary through Iterative Back-Translation,” The 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), pp.887-892 (2024).
[15] Yuki Yamamoto, Tomoyosi Akiba, and Hajime Tsukada, “Training Neural Machine Translation Models by using an Entire Comparable Corpus,” 10th International Conference on Advanced Informatics: Concepts, Theory and Applications (ICAICTA 2023) (2023).
[16] Keiya Takagi, Tomoyosi Akiba, and Hajime Tsukada, “Semi-supervised ASR based on iterative joint training with discrete speech synthesis,” 2022 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC), pp.923-928 (2022).
[17] Yuuki Yamagishi, Tomoyosi Akiba, and Hajime Tsukada, “English-Japanese machine translation for lecture subtitles using back-translation and transfer learning,” 2020 IEEE 9th Global Conference on Consumer Electronics (GCCE), pp.728-730 (2020).
[18] Takahiro Nomura, Hajime Tsukada, and Tomoyoshi Akiba, “Improvement of word alignment models for Vietnamese-to-English translation,” 12th International Workshop on Spoken Language Translation (IWSLT), pp.112-116 (2015).
[19] Katuhito Sudoh, Graham Neubig, Kevin Duh, and Hajime Tsukada, “NTT-NAIST SMT systems for IWSLT 2013,” 10th International Workshop on Spoken Language Translation (IWSLT) (2013).
[20] Katsuhiko Hayashi, Katsuhito Sudoh, Hajime Tsukada, Jun Suzuki, and Masaaki Nagata, “Shift-reduce word reordering and editing for machine translation,” Empirical Methods in Natural Language Processing (EMNLP Short Papers), pp.1382-1386 (2013).
[21] Kevin Duh, Graham Neubig, Katsuhito Sudoh, and Hajime Tsukada, “Adaptation data selection using neural language models: Experiments in machine translation,” 51st Annual Meeting of the Association for Computational Linguistics (ACL Short Papers), pp. 678-683 (2013).
[22] Katsuhito Sudoh, Jun Suzuki, Hajime Tsukada, Masaaki Nagata, Sho Hoshino, and Yusuke Miyao, “NTT-NII statistical machine translation for NTCIR-10 PatenMT,” 10th NTCIR Conference, pp.294-300 (2013).
[23] Dan Han, Katsuhito Sudoh, Xianchao Wu, Kevin Duh, Hajime Tsukada, and Masaaki Nagata, “Head finalization reordering for Chinese-to-Japanese machine translation,” 6th Workshop on Syntax, Semantics and Structure in Statistical Translation (SSST), pp.57-66 (2012).
[24] Wu Xianchao, Katsuhito Sudoh, Hajime Tsukada, and Masaaki Nagata, “A comparative study of target dependency structures for statistical machine translation,” 50th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pp.100-104 (2012).
[25] Kevin Duh, Katsuhito Sudoh, Xianchao Wu, Hajime Tsukada, and Masaaki Nagata, “Learning to translate with multiple objectives,” 50th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pp.1-10 (2012).
[26] Shuhei Kondo, Mamoru Komachi, Yuji Matsumoto, Katsuhito Sudoh, Kevin Duh, and Hajime Tsukada, “Learning of linear ordering problems and its application to J-E patent translation in NTCIR-9 PatentMT,” NTCIR-9 Workshop Meeting, pp.641-645 (2011).
[27] Katsuhito Sudoh, Xianchao Wu, Kevin Duh, Hajime Tsukada, Masaaki Nagata, Takuya Matsuzaki, and Junichi Tsujii, “NTT-UT statistical machine translation in NTCIR-9 PatentMT,” NTCIR-9 Workshop Meeting, pp.585-592 (2011).
[28] Tomoki Nagase, Hajime Tsukada, Katsunori Kotani, Nobutoshi Hatanaka, and Yoshiyuki Sakamoto, “Automatic error analysis based on grammatical questions,” 25th Pacific Asia Conference on Language, Information and Computation, pp.206-215 (2011).
[29] Xianchao Wu, Katsuhito Sudoh, Kevin Duh, Hajime Tsukada, and Masaaki Nagata, “Extracting. pre-ordering rules from predicate-argument structures,” 5th International Joint Conference on Natural Language Processing, pp.29-37 (2011).
[30] Kevin Duh, Katsuhito Sudoh, Xianchao Wu, Hajime Tsukada, and Masaaki Nagata, “Generalized minimum bayes risk system combination,” 5th International Joint Conference on Natural Language Processing, pp.1356-1360 (2011).
[31] Xianchao Wu, Katsuhito Sudoh, Kevin Duh, Hajime Tsukada, and Masaaki Nagata, “Technical report of NTT for the 7th China workshop on machine translation,” 7th China Workshop on Machine Translation (2011).
[32] Katsuhito Sudoh, Xianchao Wu, Kevin Duh, and Hajime Tsukada, “Post-ordering in statistical machine translation,” Machine Translation Summit XIII, pp.316-323 (2011).
[33] Xianchao Wu, Katsuhito Sudoh, Kevin Duh, Hajime Tsukada, and Masaaki Nagata, “Extracting pre-ordering rules from chunk-based dependency trees for Japanese-to-English translation,” Machine Translation Summit XIII, pp.300-307 (2011).
[34] Kevin Duh, Katsuhito Sudoh, Tomoharu Iwata, and Hajime Tsukada, “Alignment inference and Bayesian adaptation for machine translation,” Machine Translation Summit XIII, pp.114-121 (2011).
[35] Katsuhito Sudoh, Kevin Duh, and Hajime Tsukada, “NTT statistical machine translation system for IWSLT 2010,” 7th International Workshop on Spoken Language Translation: Evaluation Campaign, pp.147-152 (2010).
[36] Kevin Duh, Katsuhito Sudoh, and Hajime Tsukada, “Analysis of translation model adaptation in statistical machine translation,” 7th International Workshop on Spoken Language Translation, pp.243-250 (2010).
[37] Hideki Isozaki, Tsutomu Hirao, Kevin Duh, Katsuhito Sudoh, and Hajime Tsukada, “Automatic evaluation of translation quality for distant language pairs,” 2010 Conference on Empirical Methods in Natural Language Processing, pp.944-952 (2010).
[38] Katsuhiko Hayashi, Hajime Tsukada, Katsuhito Sudoh, Kevin Duh, and Seiichi Yamamoto, “Hierarchical phrase-based machine translation with word-based reordering model,” 23rd International Conference on Computational Linguistics (Coling 2010), pp.439-446 (2010).
[39] Katsuhito Sudoh, Kevin Duh, Hajime Tsukada, Tsutomu Hirao, and Masaaki Nagata, “Divide and translate: Improving long distance reordering in statistical machine translation,” Joint Fifth Workshop on Statistical Machine Translation and MetricsMATR, pp.418-427 (2010).
[40] Hideki Isozaki, Katsuhito Sudoh, Hajime Tsukada, and Kevin Duh, “Head finalization: A simple. reordering rule for SOV languages,” Joint Fifth Workshop on Statistical Machine Translation and MetricsMATR, pp.244-251 (2010).
[41] Kevin Duh, Katsuhito Sudoh, Hajime Tsukada, Hideki Isozaki, “N-best reranking by multitask learning,” Joint Fifth Workshop on Statistical Machine Translation and MetricsMATR, pp.375-383 (2010).

[42] Shin'ya Nakajima and Hajime Tsukada, “Prosodic Features of Utterances in Task-Oriented Dialogues,” Computing Prosody, Yoshinori Sagisaka, Nick Campbell, and Norio Higuchi eds., pp.81-93, Springer-Verlag (1997).
[43] 塚田 元, “3.1.2節 統計に基づく翻訳,” 言語処理学事典, 石崎 俊 編, pp.266-269, 共立出版 (2009).

1. 言語処理学会論文賞: 林 克彦, 須藤 克仁, 塚田 元, 鈴木 潤, 永田 昌明, “単語並べ替えと冠詞生成の同時逐次処理: 日英機械翻訳への適用,” 言語処理学会 (2014).
2. 優秀発表賞: 平尾 努, 磯崎 秀樹, Kevin Duh, 須藤 克仁, 塚田 元, 永田 昌明, “RIBES:順位相関に基づく翻訳の自動評価法,” 言語処理学会 (2011).
3. アジア言語処理優秀論文賞: Katsuhito Sudoh, Hajime Tsukada, Hideki Isozaki, 21st International Conference on Computational Linguistics and 44th Annual Meeting of the Association for Computational Linguistics (COLING-ACL 2006) (2006).

2号館4階 2410室