Papers by Masaaki Nagata

Pruning Basic Elements for Better Automatic Evaluation of Summaries (N18-2)

Challenge:	Recent machine translation algorithms rely on parallel corpora, but only some resource-rich language pairs can benefit from them.
Approach:	They construct a parallel corpus for English-Japanese, which has 8.7 million sentence pairs . they use a web crawler to automatically align parallel sentences in the corpus .
Outcome:	The proposed corpus includes a broader range of domains and can be trained with a pre-trained model.

Copied to clipboard

Ukyo Honda, Tsutomu Hirao, Masaaki Nagata

Summarization

Challenge:	Summarization studies work on increasing the scores that are given by automatic evaluation measures.
Approach:	They propose a simple but highly effective automatic evaluation measure of summarization, pruned Basic Elements.
Outcome:	The proposed measure outperforms ROUGE and BE in most cases and achieves highest correlation coefficient in TAC 2011 AESOP task.

Improving Neural Machine Translation by Incorporating Hierarchical Subword Features (C18-1)

Copied to clipboard

NeoAMT: Neologism-Aware Agentic Machine Translation with Reinforcement Learning (2026.acl-long)

Challenge:	Using subwords, we find that the appropriate subword units for the three layers differ depending on the model . incorporating hierarchical subword features improves BLEU scores on the IWSLT evaluation datasets.
Approach:	They propose a method that expresses a word by combining "subwords" they propose to incorporate hierarchical subword features into a single embedding layer .
Outcome:	The proposed method improves BLEU scores on the IWSLT evaluation datasets.

Copied to clipboard

Zhongtao Miao, Kaiyan Zhao, Masaaki Nagata, Yoshimasa Tsuruoka

Machine Translation

Challenge:	Neologism-aware machine translation aims to translate source sentences containing neologismes into target languages.
Approach:	They propose an agentic framework for neologism-aware machine translation equipped with a Wiktionary-based search toolkit.
Outcome:	The proposed framework is based on a Wiktionary-based search toolkit and a dedicated dataset for neologism-aware machine translation.

A Simple and Strong Baseline for End-to-End Neural RST-style Discourse Parsing (2022.findings-emnlp)

Copied to clipboard

Naoki Kobayashi, Tsutomu Hirao, Hidetaka Kamigaito, Manabu Okumura, Masaaki Nagata

NTT Neural Machine Translation Systems at WAT 2019 (D19-52)

Challenge:	Existing discourse parsing methods need a strong baseline for reporting reliable experimental results.
Approach:	They integrate existing parsing strategies with transformer-based pre-trained language models to provide a strong baseline for reporting reliable experimental results.
Outcome:	The proposed model outperforms the current best model using DeBERTa.

Copied to clipboard

Word Alignment as Preference for Machine Translation (2024.emnlp-main)

Challenge:	We submitted two systems for scientific paper subtask and timely disclosure subtask . we evaluated the usefulness of incorporating external data from a wide variety of web pages to improve the translation quality.
Approach:	They describe two different translation tasks submitted to WAT 2019 . they submitted scientific paper subtasks and timely disclosure subtask .
Outcome:	The proposed system performed better on scientific paper and timely disclosure subtasks.

Copied to clipboard

Qiyu Wu, Masaaki Nagata, Zhongtao Miao, Yoshimasa Tsuruoka

Machine Translation

Challenge:	Hallucination and omission are a problem in machine translation because of an LLM's size and low-resource languages.
Approach:	They propose to use word alignment as preference to optimize an LLM-based MT model to mitigate hallucination and omission problems.
Outcome:	The proposed model is able to mitigate hallucination and omission by using word alignment as preference.

Split or Merge: Which is Better for Unsupervised RST Parsing? (D19-1)

Copied to clipboard

Naoki Kobayashi, Tsutomu Hirao, Kengo Nakamura, Hidetaka Kamigaito, Manabu Okumura, Masaaki Nagata

Responsible & Trustworthy NLP

Challenge:	Rhetorical Structure Theory (RST) parsers have been based on supervised learning approaches that require an annotated corpus of sufficient size and quality.
Approach:	They propose two unsupervised methods that build an optimal RST tree based on a dissimilarity score function for splitting a text span into smaller ones and a similarity score for merging two adjacent spans into a large one.
Outcome:	The proposed method achieves the best score on English and German RST treebanks, around 0.8 F1 score, close to the previous supervised parsers.

An Empirical Study of Building a Strong Baseline for Constituency Parsing (P18-2)

Copied to clipboard

Jun Suzuki, Sho Takase, Hidetaka Kamigaito, Makoto Morishita, Masaaki Nagata

A Test Set for Discourse Translation from Japanese to English (2020.lrec-1)

Challenge:	Sequence-to-sequence models have been used for natural language generation tasks such as machine translation and summarization.
Approach:	They propose to build a strong baseline based on general purpose sequence-to-sequence models for constituency parsing.
Outcome:	The proposed model outperforms existing models in natural language generation tasks without any explicit task-specific knowledge or architecture of constituent parsing.

Copied to clipboard

Masaaki Nagata, Makoto Morishita

Context-aware Neural Machine Translation with Coreference Information (D19-65)

Challenge:	Compared with a previous study on test sets for English-to-French discourse translation, we needed different approaches because Japanese has zero pronouns and represents different senses in different characters.
Approach:	They used a test set for Japanese-to-English discourse translation to evaluate the power of context-aware machine translation.
Outcome:	The results show that the translation accuracy of Japanese-to-English discourse translation is improved by using context-aware neural machine translation.

Copied to clipboard

Takumi Ohtani, Hidetaka Kamigaito, Masaaki Nagata, Manabu Okumura

Zero Pronouns Identification based on Span prediction (2021.acl-srw)

Challenge:	Existing models for translating a sentence in a text do not consider coreference relations provided within the text.
Approach:	They propose a graph-based encoder which can consider coreference relations provided within the text explicitly.
Outcome:	The proposed model improves on the previous approach by 0.9 points on the BLEU score . the graph-based encoder can handle a longer text well, compared with the previous model .

Copied to clipboard

Sei Iwata, Taro Watanabe, Masaaki Nagata

Challenge:	Pro-drop languages allow omissions of essential phrases or arguments . the presence of zero-pronouns affects downstream tasks of NLP .
Approach:	They propose a query-based method to identify zero-pronoun arguments . they use Japanese and Chinese datasets to evaluate the method .
Outcome:	The proposed method surpasses the sequence labeling baseline on Japanese and Chinese datasets.

Mixed Multi-Head Self-Attention for Neural Machine Translation (D19-56)

Copied to clipboard

Hongyi Cui, Shohei Iida, Po-Hsuan Hung, Takehito Utsuro, Masaaki Nagata

Neural Tensor Networks with Diagonal Slice Matrices (N18-1)

Challenge:	Recent advances in neural machine translation have been made in the field of multi-head self-attention and there is no explicit mechanism to ensure that different attention heads capture different features.
Approach:	They propose a novel multi-head self-attention model which models not only global and local attention but also forward and backward attention in different attention heads.
Outcome:	The proposed model improves on WAT17 English-Japanese and IWSLT14 German-English translation tasks without increasing the number of parameters.

Copied to clipboard

Takahiro Ishihara, Katsuhiko Hayashi, Hitoshi Manabe, Masashi Shimbo, Masaaki Nagata

Knowledge Representation

Challenge:	A large number of parameters can cause overfitting and a long training time for neural tensor networks (NTNs).
Approach:	They propose two new parameter reduction techniques to reduce the number of parameters in an NTN without diminishing its expressiveness.
Outcome:	The proposed models learn better and faster than the original (R)NTNs.

Automatic Pyramid Evaluation Exploiting EDU-based Extractive Reference Summaries (D18-1)

Copied to clipboard

Tsutomu Hirao, Hidetaka Kamigaito, Masaaki Nagata

Challenge:	Existing methods for evaluating content are not accurate because they only confirm if the summary contains small textual fragments.
Approach:	They propose to transform human-made reference summaries into extractive reference sums and weight them using elementary discourse units.
Outcome:	The proposed method strongly correlates with manual evaluations on DUC and TAC data sets.

Provable Fast Greedy Compressive Summarization with Any Monotone Submodular Function (N18-1)

Copied to clipboard

Shinsaku Sakaue, Tsutomu Hirao, Masaaki Nishino, Masaaki Nagata

Document Alignment based on Overlapping Fixed-Length Segments (2024.acl-srw)

Challenge:	Submodular maximization with the greedy algorithm is an effective approach to extractive summarization.
Approach:	They propose a submodular maximization method that is 100 to 400 times faster than existing methods for extractive summarization.
Outcome:	The proposed method is 100 to 400 times faster than existing method based on integer-linear-programming formulations and achieves 95%-approximation.

Copied to clipboard

Xiaotian Wang, Takehito Utsuro, Masaaki Nagata

Challenge:	Existing studies show that web crawling can be used to obtain large-scale parallel corpora for NLP tasks.
Approach:	They propose a sentence-based segmentation method for document alignment . they compare it with a fixed-length segmentation technique to handle long-text encoding better.
Outcome:	The proposed method improves document alignment and recall by 1% to 10% on a document alignment task for Japanese-English and French-English datasets.

Sequential Span Classification with Neural Semi-Markov CRFs for Biomedical Abstracts (2020.findings-emnlp)

Copied to clipboard

Kosuke Yamada, Tsutomu Hirao, Ryohei Sasano, Koichi Takeda, Masaaki Nagata

Text Classification

Challenge:	Existing methods for dividing biomedical abstracts into rhetorical segments assign a rhetorical label to each sentence while considering context in the abstract.
Approach:	They propose to use Neural Semi-Markov Conditional Random Fields to assign a rhetorical label to a span that consists of continuous sentences.
Outcome:	The proposed method achieved the best micro sentence-F1 score and the best macro span-F1.

JaParaPat: A Large-Scale Japanese-English Parallel Patent Application Corpus (2024.lrec-main)

Copied to clipboard

Masaaki Nagata, Makoto Morishita, Katsuki Chousa, Norihito Yasuda

Answering while Summarizing: Multi-task Learning for Multi-hop QA with Evidence Extraction (P19-1)

Challenge:	a recent study has demonstrated that patent translation accuracy improves as the amount of training data or the number of model parameters increases.
Approach:	They construct a bilingual corpus of Japanese-English patent application data from 2000 to 2021 . they extracted 1.4M Japanese- English document pairs and extracted 350M sentence pairs .
Outcome:	The proposed method improves translation accuracy by 20 bleu points . it is the first publicly available large-scale Japanese-English patent corpus .

Copied to clipboard

Kosuke Nishida, Kyosuke Nishida, Masaaki Nagata, Atsushi Otsuka, Itsumi Saito, Hisako Asano, Junji Tomita

Information Extraction & Text Mining

Challenge:	Question answering (QA) using textual sources for purposes such as reading comprehension has attracted much attention.
Approach:	They propose a Query Focused Extractor model for evidence extraction and multi-task learning with the QA model.
Outcome:	The proposed model achieves state-of-the-art evidence extraction score on hotpotQA and FEVER, which is a recognizing textual entailment task on a large textual database.

Improving Neural RST Parsing Model with Silver Agreement Subtrees (2021.naacl-main)

Copied to clipboard

Naoki Kobayashi, Tsutomu Hirao, Hidetaka Kamigaito, Manabu Okumura, Masaaki Nagata

Challenge:	Existing methods for Rhetorical Structure Theory (RST) parsing use supervised learning, but the RST-DT is small due to the costly annotation of RST trees.
Approach:	They propose to use silver data to improve RST parsing models by using annotated silver data.
Outcome:	The proposed method achieves the best micro-F1 scores for Nuclearity and Relation at 75.0 and 63.2 . it also achieves a remarkable gain in relation score against the previous state-of-the-art parser.

Using Semantic Similarity as Reward for Reinforcement Learning in Sentence Generation (P19-2)

Copied to clipboard

Go Yasui, Yoshimasa Tsuruoka, Masaaki Nagata

Context-aware Neural Machine Translation with Mini-batch Embedding (2021.eacl-main)

Challenge:	Existing models for sentence generation use cross-entropy loss as the loss function . however, cross-etropy is unable to evaluate sentences as a whole and lacks flexibility . et al., 2018: a novel approach to improve sentence generation models .
Approach:	They propose a method to train a model using estimated semantic similarity between output and reference sentences to alleviate cross-entropy loss problems.
Outcome:	The proposed model improves the BLEU scores from the baseline LSTM NMT model.

Copied to clipboard

Makoto Morishita, Jun Suzuki, Tomoharu Iwata, Masaaki Nagata

Representation Learning

Challenge:	Existing models that translate sentences in sentence-by-sentence manner do not consider the inter-sentent context in a language translation model.
Approach:	They propose to use mini-batch embedding to represent the features of sentences in a mini-Batch by choosing sentences from the same document and then embedd it into the model.
Outcome:	The proposed method outperforms baseline models and improves writing style or terminology to fit the document’s context.

Generating Natural Anagrams: Towards Language Generation Under Hard Combinatorial Constraints (D19-1)

Copied to clipboard

Masaaki Nishino, Sho Takase, Tsutomu Hirao, Masaaki Nagata

Direct Output Connection for a High-Rank Language Model (D18-1)

Challenge:	Existing methods for creating anagrams do not pay much attention to the naturalness of the generated anagramms.
Approach:	They propose to combine depth-first search with modern neural language models to generate anagrams by permutation of characters in an input sentence or phrase.
Outcome:	The proposed method generates significantly more natural anagrams than baseline methods.

Copied to clipboard

Sho Takase, Jun Suzuki, Masaaki Nagata

Argument Mining as a Text-to-Text Generation Task (2024.eacl-long)

Challenge:	Neural network language models have played a central role in recent natural language processing advances.
Approach:	They propose a recurrent neural network language model that combines probability distributions computed from a final RNN layer and also middle layers.
Outcome:	The proposed language model achieves the best score on the Penn Treebank and WikiText-2 benchmark datasets.

Copied to clipboard

Masayuki Kawarada, Tsutomu Hirao, Wataru Uchida, Masaaki Nagata

Attention over Heads: A Multi-Hop Attention for Neural Machine Translation (P19-2)

Challenge:	Argument Mining (AM) aims to uncover the argumentative structures within a text.
Approach:	They propose a method that generates argumentatively annotated text using a pretrained encoder-decoder language model and a pre-trained decoder.
Outcome:	The proposed method achieves state-of-the-art performance on three types of benchmark datasets.

Copied to clipboard

Shohei Iida, Ryuichiro Kimura, Hongyi Cui, Po-Hsuan Hung, Takehito Utsuro, Masaaki Nagata

WSPAlign: Word Alignment Pre-training via Large-Scale Weakly Supervised Span Prediction (2023.acl-long)

Challenge:	Existing multihop attentions for machine comprehension are recurrent and hierarchical . a proposed multi-hop attention for the Transformer refines the attention for an output symbol many times .
Approach:	They propose a multi-hop attention for the Transformer which integrates attentions from each head.
Outcome:	The proposed model outperforms the baseline Transformer in terms of translation accuracy and speed.

Copied to clipboard

Qiyu Wu, Masaaki Nagata, Yoshimasa Tsuruoka

SpanAlign: Sentence Alignment Method based on Cross-Language Span Prediction and ILP (2020.coling-main)

Challenge:	Existing word alignment methods rely on manual data and lack generalization ability.
Approach:	They propose to use a weakly-supervised large-scale weakly supervised dataset for word alignment pre-training via span prediction to reduce the need for manual data.
Outcome:	The proposed method improves upon the best supervised baseline by 3.3 6.1 points in F1 and 1.5 6.1 point in AER.

Copied to clipboard

Katsuki Chousa, Masaaki Nagata, Masaaki Nishino

Higher-Order Syntactic Attention Network for Longer Sentence Compression (N18-1)

Challenge:	Existing methods for automatic sentence alignment assume monotonic alignments, but they can handle non-monotonic alignments.
Approach:	They propose a method to automatically extract parallel sentences from noisy parallel documents by embeddings and encoding each source and target sentence.
Outcome:	The proposed method improves translation accuracy by 4.1 BLEU scores on English-Japanese . it can predict spans in target document from sentences in source document .

Copied to clipboard

Hidetaka Kamigaito, Katsuhiko Hayashi, Tsutomu Hirao, Masaaki Nagata

BiMax: Bidirectional MaxSim Score for Document-Level Alignment (2025.findings-emnlp)

Challenge:	Existing sentence compression methods do not handle syntactic features, causing performance degradation . et al. (2015) reported that the longer the input sentences are, the worse the performance becomes.
Approach:	They propose a higher-order syntactic attention network that handles higher-level dependency features as an attention distribution on LSTM hidden states.
Outcome:	The proposed method outperforms baseline methods on a Google sentence compression dataset.

Copied to clipboard

Xiaotian Wang, Takehito Utsuro, Masaaki Nagata

Representation Learning

Challenge:	Document alignment is necessary for the hierarchical mining of documents across source and target languages.
Approach:	They propose a cross-lingual Bidirectional Maxsim score for computing doc-to-doc similarity.
Outcome:	The proposed method achieves accuracy comparable to OT with an approximate 100-fold speed increase.

Domain Adaptation of Machine Translation with Crowdworkers (2022.emnlp-industry)

Copied to clipboard

JParaCrawl v3.0: A Large-scale English-Japanese Parallel Corpus (2022.lrec-1)

Challenge:	Existing methods for machine translation work well when no in-domain data are available.
Approach:	They propose a framework that efficiently and effectively collects parallel sentences in a target domain from the web with the help of crowdworkers.
Outcome:	The proposed method improves the BLEU scores by an average of +7.8 points compared to a general-purpose translation model.

Copied to clipboard

Makoto Morishita, Katsuki Chousa, Jun Suzuki, Masaaki Nagata

Machine Translation

Challenge:	Existing parallel corpora for English-Japanese are limited, limiting the accuracy of machine translation models.
Approach:	They propose a web-based English-Japanese parallel corpus with 21 million unique sentence pairs . this is more than twice as many as the previous corpus JParaCrawl v2.0 .
Outcome:	The proposed corpus boosts the accuracy of machine translation models on various domains.

Case-Based Decision-Theoretic Decoding with Quality Memories (2025.emnlp-main)

Copied to clipboard

Hiroyuki Deguchi, Masaaki Nagata

A Supervised Word Alignment Method based on Cross-Language Span Prediction using Multilingual BERT (2020.emnlp-main)

Challenge:	Minimum Bayes risk (MBR) decoding is a decision rule of text generation . however, it depends on sample texts drawn from the text generation model .
Approach:	They propose a case-based decision-theoretic method to estimate the expected utility using examples of domain data.
Outcome:	The proposed method outperforms MAP decoding in translation tasks and image captioning tasks on MSCOCO and nocaps datasets.

Copied to clipboard

Masaaki Nagata, Katsuki Chousa, Masaaki Nishino

Improving Word Alignment Using Semi-Supervised Learning (2025.findings-acl)

Challenge:	supervised word alignment tools such as GIZA++, MGIZA (Gao and Vogel, 2008) and FastAlign remain stagnant in terms of word alignment accuracy.
Approach:	They propose a supervised word alignment method based on cross-language span prediction by formalizing a word alignment problem as a collection of independent predictions from a token in the source sentence to a span in the target sentence.
Outcome:	The proposed method significantly outperforms previous supervised and unsupervised word alignment methods without any bitexts for pretraining.

Copied to clipboard

Zhongtao Miao, Qiyu Wu, Masaaki Nagata, Yoshimasa Tsuruoka