Papers by Yuji Matsumoto
Ranking-Based Automatic Seed Selection and Noise Reduction for Weakly Supervised Relation Extraction (P18-2)
Copied to clipboard
| Challenge: | et al., 1998: bootstrapping for relation extraction uses minimally supervised methods . etudes show that proposed methods for automatic seed selection and noise reduction are better than baseline systems . |
| Approach: | They propose automatic seed selection and noise reduction for distantly supervised relation extraction tasks. |
| Outcome: | The proposed methods achieve better performance than baseline systems in both tasks. |
A Parallel Corpus of Arabic-Japanese News Articles (L18-1)
Copied to clipboard
| Challenge: | a large-scale parallel corpora with manually verified subsets of sentences has been used for machine translation between major language pairs. |
| Approach: | They describe the creation process and statistics of the Arabic-Japanese portion of the TUFS Media Corpus . they also report the first results of Arabic-japanese phrase-based machine translation trained on the corpus based on the Arabic corpus. |
| Outcome: | The proposed corpus is a document-level parallel corpus and sentence-level parser corpus . it is the first time that Arabic-Japanese translations have been trained on it . |
Coordination Generation via Synchronized Text-Infilling (2022.coling-1)
Copied to clipboard
| Challenge: | Generating synthetic data from pre-trained language models has enhanced performance across several NLP tasks. |
| Approach: | They propose a method for generating sentences with a coordinate structure in which the boundaries of its conjuncts are explicitly specified. |
| Outcome: | The proposed method produces promising coordination instances that provide gains for the task in low-resource settings. |
Unsupervised Multilingual Word Embedding with Limited Resources using Neural Language Models (P19-1)
Copied to clipboard
| Challenge: | Existing methods that map word embeddings into a common space without any parallel data or pre-training have been proposed that are limited in resources and perform poorly under resource-poor conditions. |
| Approach: | They propose a model that maps monolingual word embeddings into a common space without any parallel data and generates multilingual embeddables without any pre-training. |
| Outcome: | The proposed model outperforms existing methods on word alignment tasks on low-resource conditions and with limited resources. |
Unsupervised Paraphrasing of Multiword Expressions (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for paraphrasing multiword expressions in context are unsupervised . multiwords are notoriously difficult to model because the meaning of the whole can diverge substantially from that of the component words. |
| Approach: | They propose an unsupervised approach to paraphrasing multiword expressions in context using monolingual corpus data and pre-trained language models. |
| Outcome: | The proposed method outperforms all unsupervised systems and rivals supervised systems on the SemEval 2022 idiomatic text similarity task. |
Recent Trends in Personalized Dialogue Generation: A Review of Datasets, Methodologies, and Evaluations (2024.lrec-main)
Copied to clipboard
| Challenge: | Personalization is a multifaceted process that requires multiple definitions and varies between individuals. |
| Approach: | They propose to systemically survey the recent landscape of personalized dialogue generation including the datasets employed, methodologies developed, and evaluation metrics applied. |
| Outcome: | The proposed model can generate fluent and coherent responses to human queries in a language-based conversational agent. |
Dissecting GraphRAG: A Modular Analysis of Knowledge Structuring for Factoid Question Answering (2026.tacl-1)
Copied to clipboard
Noriki Nishida, Rumana Ferdous Munne, Shanshan Liu, Narumi Tokunaga, Yuki Yamagata, Fei Cheng, Kouji Kozaki, Yuji Matsumoto
| Challenge: | GraphRAG integrates structured knowledge graphs into question answering . high-quality triple extraction is critical, but lacks granularity and topical coherence . large language models suffer from inherent limitations in their internalized knowledge . |
| Approach: | They evaluate module-level design choices in GraphRAG for retrieval-augmented generation . they find that triple extraction is critical for accurate and comprehensive retrieval . |
| Outcome: | The proposed framework outperforms other retrieval-augmented generation frameworks in accuracy and efficiency. |
PolyMinder: A Support System for Entity Annotation and Relation Extraction in Polymer Science Documents (2025.coling-demos)
Copied to clipboard
| Challenge: | Automated Named Entity Recognition (NER) and Relation Extraction (RE) models are tailored to the polymer domain. |
| Approach: | They propose to automate the annotation process by providing a web-based interface where users can visualize, verify, and refine the extracted information before finalizing the annotations. |
| Outcome: | The proposed system streamlines the annotation process by providing a web-based interface where users can visualize, verify, and refine the extracted information before finalizing the annotations. |
CovRelex-SE: Adding Semantic Information for Relation Search via Sequence Embedding (2023.eacl-demo)
Copied to clipboard
| Challenge: | COVID-19 has affected all aspects of human life, causing problems related to acronyms, synonyms, and rare keywords. |
| Approach: | They propose a hybrid relation retrieval system based on embeddings to provide high-quality search results. |
| Outcome: | The proposed system can be accessed through the following URL: http://www.jaist.ac.jp/is/labs/nguyen-lab/systems/covrelex-se/. |
Universal Dependencies Version 2 for Japanese (L18-1)
Copied to clipboard
Masayuki Asahara, Hiroshi Kanayama, Takaaki Tanaka, Yusuke Miyao, Sumire Uematsu, Shinsuke Mori, Yuji Matsumoto, Mai Omura, Yugo Murawaki
| Challenge: | UD Japanese resources are built on automatic conversion from several treebanks. |
| Approach: | They propose to port the word delimitation, POS, and syntactic relations of existing treebanks to UD Japanese . they discuss the issues of the UD scheme found through porting of the Japanese language . |
| Outcome: | The proposed UD Japanese resources are based on automatic conversion from treebanks. |
Zero-Shot Entailment Learning for Ontology-Based Biomedical Annotation Without Explicit Mentions (2025.coling-main)
Copied to clipboard
Rumana Ferdous Munne, Noriki Nishida, Shanshan Liu, Narumi Tokunaga, Yuki Yamagata, Kouji Kozaki, Yuji Matsumoto
| Challenge: | Automated biomedical annotation presents significant challenges when entities are not explicitly mentioned in the text. |
| Approach: | They propose an entailment-based zero-shot text classification approach to annotate biomedical text passages using the Homeostasis Imbalance Process (HOIP) ontology. |
| Outcome: | The proposed method performs well when processes are not explicitly mentioned . it is time-consuming and expensive to annotate biomedical texts with a specific ontology . |
A Span Selection Model for Semantic Role Labeling (D18-1)
Copied to clipboard
| Challenge: | Existing models for semantic role labeling use BIO tags to predict argument spans . but performance of these approaches is weak . |
| Approach: | They propose a span-based model that takes into account all possible argument spans and scores them for each label. |
| Outcome: | The proposed model achieves state-of-the-art results on the CoNLL-2005 and 2012 datasets. |
Dynamic Feature Selection with Attention in Incremental Parsing (C18-1)
Copied to clipboard
| Challenge: | Currently, incremental transition-based parsers require that all inputs are visible from the beginning to extract good features from a limited local context. |
| Approach: | They propose a technique to maximize local features with an attention mechanism which works as context- dependent dynamic feature selection. |
| Outcome: | The proposed technique can extract features from a limited local context and is able to perform multilingual experiments and demon strate on local ambiguous points. |
MA-COIR: Leveraging Semantic Search Index and Generative Models for Ontology-Driven Biomedical Concept Recognition (2025.acl-srw)
Copied to clipboard
Shanshan Liu, Noriki Nishida, Rumana Ferdous Munne, Narumi Tokunaga, Yuki Yamagata, Kouji Kozaki, Yuji Matsumoto
| Challenge: | Existing concepts recognition methods that rely on explicit mention identification fail to capture complex concepts not explicitly stated in the text. |
| Approach: | They propose a framework that reformulates concept recognition as an indexing-recognition task. |
| Outcome: | The proposed framework reduces computational requirements and improves recognition efficiency in low-resource settings. |
Unsupervised Lexical Substitution with Decontextualised Embeddings (2022.coling-1)
Copied to clipboard
| Challenge: | Existing methods for lexical substitution using pre-trained language models have some limitations. |
| Approach: | They propose an unsupervised method for lexical substitution using pre-trained language models. |
| Outcome: | The proposed method outperforms baseline models and establishes a state-of-the-art without supervision or fine-tuning. |
Out-of-Domain Discourse Dependency Parsing via Bootstrapping: An Empirical Analysis on Its Effectiveness and Limitation (2022.tacl-1)
Copied to clipboard
| Challenge: | Discourse parsing accuracy degrades significantly on out-of-domain text. |
| Approach: | They propose to use bootstrapping methods to adapt modern discourse dependency parsers to out-of-domain text without additional human supervision. |
| Outcome: | The proposed methods are significantly and consistently effective for unsupervised domain adaptation of discourse dependency parsing, but the low coverage of accurately predicted pseudo labels is a bottleneck for further improvement. |
A Unified Framework for N-ary Property Information Extraction in Materials Science (2025.findings-emnlp)
Copied to clipboard
| Challenge: | a framework for extracting n-ary property information from materials science literature is proposed . the framework addresses the critical challenge of capturing complex relationships that span multiple sentences. |
| Approach: | They propose a framework for extracting n-ary property information from materials science literature . they propose three complementary approaches to capture complex relationships that span multiple sentences . |
| Outcome: | The proposed framework outperforms existing methods in n-ary property extraction tasks. |
EMTC: Multilabel Corpus in Movie Domain for Emotion Analysis in Conversational Text (L18-1)
Copied to clipboard
| Challenge: | Existing emotion corpora collected from twitters and use hashtags are limited in the number of characters. |
| Approach: | They propose to build an emotion corpus based on conversational text data that includes 2.1 million utterances and is partly annotated by ourselves and independent annotators. |
| Outcome: | The proposed corpus includes conversations from movies with more than 2.1 million utterances which are partly annotated by ourselves and independent annotators. |
CovRelex: A COVID-19 Retrieval System with Relation Extraction (2021.eacl-demos)
Copied to clipboard
| Challenge: | Existing challenges to making the system more practical include dealing with newly created and unknown data, and solving the performance gap when utilizing present data. |
| Approach: | They propose a scientific paper retrieval system targeting entities and relations via relation extraction on COVID-19 scientific papers. |
| Outcome: | The proposed system can be accessed via https://www.jaist.ac.jp/is/labs/nguyen-lab/systems/covrelex/. |
Removing Word-Level Spurious Alignment between Images and Pseudo-Captions in Unsupervised Image Captioning (2021.eacl-main)
Copied to clipboard
| Challenge: | Unsupervised image captioning is a challenging task that requires manual annotation. |
| Approach: | They propose a simple gating mechanism that is trained to align image features with the most reliable words in pseudo-captions. |
| Outcome: | The proposed method outperforms the previous methods without complex learning objectives. |
Decomposed Local Models for Coordinate Structure Parsing (N19-1)
Copied to clipboard
| Challenge: | Existing methods for coordination boundary identification are inefficient, even for humans. |
| Approach: | They propose a simple and accurate model for coordination boundary identification . they combine syntactic parsers and neural networks to compute similarity and replaceability features of conjuncts . |
| Outcome: | The proposed model outperforms similarity-based approaches but cannot handle more than two conjuncts in a coordination and multiple coordinations at once. |
PolyNERE: A Novel Ontology and Corpus for Named Entity Recognition and Relation Extraction in Polymer Science Domain (2024.lrec-main)
Copied to clipboard
| Challenge: | a new ontology for polymer-relevant entities and relations is available for training data . the ontologies are customizable to adapt to specific research needs. |
| Approach: | They propose a polymer-relevant ontology featuring crucial entities and relations . the ontologies are customizable to adapt to specific research needs . |
| Outcome: | The proposed ontology can extract polymer-relevant information from scientific papers . it can be customized to adapt to specific research needs . |
Dependency Patterns of Complex Sentences and Semantic Disambiguation for Abstract Meaning Representation Parsing (2021.starsem-1)
Copied to clipboard
| Challenge: | Abstract Meaning Representation (AMR) is a sentence-level meaning representation based on predicate argument structure. |
| Approach: | They propose to use a dictionary to capture the structure of complex sentences . they train models on data derived from AMR and Wikipedia corpus . |
| Outcome: | The proposed model will be made public and the proposed patterns will be validated. |
Chemical Compounds Knowledge Visualization with Natural Language Processing and Linked Data (L18-1)
Copied to clipboard
| Challenge: | Existing systems for chemical compounds extraction and registration depend on human labor . CAS databases are being created, but information written in other languages is not exploited well . |
| Approach: | They propose a visualization system for chemical compounds extracted from Japanese texts and chemical compound databases represented as Linked Data (LD) system integrates extracted results with existing chemical compound knowledge to provide different views of chemical compounds. |
| Outcome: | The proposed system integrates extraction results with existing chemical compound knowledge to provide different views of chemical compounds. |
Wikipedia2Vec: An Efficient Toolkit for Learning and Visualizing the Embeddings of Words and Entities from Wikipedia (2020.emnlp-demos)
Copied to clipboard
Ikuya Yamada, Akari Asai, Jin Sakuma, Hiroyuki Shindo, Hideaki Takeda, Yoshiyasu Takefuji, Yuji Matsumoto
| Challenge: | Existing tools for learning the embeddings of words and entities from Wikipedia are not yet available. |
| Approach: | They propose a Python-based tool for learning Wikipedia embeddings from Wikipedia . they use a Wikipedia dump file as an argument to issue a single command . |
| Outcome: | The proposed tool achieves state-of-the-art results on the KORE entity relatedness dataset and competitive results on benchmark datasets. |
Applicability Condition Extraction for Therapeutic Drug-Disease Relations (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for identifying conditions under which a drug can be effective are limited . et al., j. n. d., al. c., and dr. m. s., 2005, are not able to identify context-specific conditions for therapeutic drug–disease relations. |
| Approach: | They propose to annotate triples of drugs, diseases, and applicability conditions from biomedical literature. |
| Outcome: | The proposed method outperforms baselines across evaluation settings. |
Nested Named Entity Recognition via Explicitly Excluding the Influence of the Best Path (2021.acl-long)
Copied to clipboard
| Challenge: | Existing methods for named entity recognition assume entities are not nested within other entities, so-called flat NER. |
| Approach: | They propose a layered method for nested named entity recognition . they use a set of hidden states to exclude the influence of the best path . |
| Outcome: | The proposed method performs better on ACE2004, ACE2005, and GENIA datasets. |
Relation Classification Using Segment-Level Attention-based CNN and Dependency-based RNN (N19-1)
Copied to clipboard
| Challenge: | Recent work on relation classification has gained much success by exploiting deep neural networks. |
| Approach: | They propose a relation classification model using Segment-level Attention-based Convolutional Neural Networks and Dependency-based Recurrent Neural networks. |
| Outcome: | The proposed model is comparable to the state-of-the-art without external lexical features on the SemEval-2010 dataset. |
LUKE: Deep Contextualized Entity Representations with Entity-aware Self-attention (2020.emnlp-main)
Copied to clipboard
| Challenge: | Existing models for entity representations do not capture information in a knowledge base, and cannot represent entities that do not exist in the KB. |
| Approach: | They propose a pretrained contextualized representation of words and entities based on the bidirectional transformer. |
| Outcome: | The proposed model achieves impressive empirical performance on a wide range of entity-related tasks. |
Better Generalizing to Unseen Concepts: An Evaluation Framework and An LLM-Based Auto-Labeled Pipeline for Biomedical Concept Recognition (2026.eacl-long)
Copied to clipboard
Shanshan Liu, Noriki Nishida, Fei Cheng, Narumi Tokunaga, Rumana Ferdous Munne, Yuki Yamagata, Kouji Kozaki, Takehito Utsuro, Yuji Matsumoto
| Challenge: | Existing methods for recognizing ontology concepts are limited by the number of annotations available. |
| Approach: | They propose an evaluation framework built on hierarchical concept indices and novel metrics to measure generalization. |
| Outcome: | The proposed evaluation framework is built on hierarchical concept indices and novel metrics to measure generalization. |
Sentence Suggestion of Japanese Functional Expressions for Chinese-speaking Learners (P18-4)
Copied to clipboard
| Challenge: | a large number of Chinese characters are commonly used both in Chinese and Japanese. |
| Approach: | They propose a computer-assisted learning system for Chinese-speaking learners of Japanese as a second language (JSL) they use a free Japanese morphological analyzer MeCab to learn Japanese functional expressions with suggestion of appropriate example sentences. |
| Outcome: | The proposed system automatically recognizes Japanese functional expressions using a free Japanese morphological analyzer and is retrained on a new conditional random field model. |
Stochastic Tokenization with a Language Model for Neural Text Classification (P19-1)
Copied to clipboard
| Challenge: | Sentences segmented with words or subwords can be difficult to perform text classification tasks. |
| Approach: | They propose a method to learn tokenization and text classification simultaneously to address these problems. |
| Outcome: | The proposed method improves on sentiment analysis in Japanese and Chinese using tokenization and text classification models. |
Structured Refinement for Sequential Labeling (2021.findings-acl)
Copied to clipboard
| Challenge: | Existing work on identifying target-irrelevant information relies on locally normalized attention without considering possible labels at other time steps. |
| Approach: | They propose to extend local normalized attention to leverage structural information for refinement . they propose to use two implementation tricks to accelerate CRF computation and an initialization trick for Chinese character embeddings . |
| Outcome: | The proposed method can be extended to include Chinese character embeddings and two implementation tricks to accelerate CRF computation. |
JaCorpTrack: Corporate History Event Extraction for Tracking Organizational Changes (2025.emnlp-industry)
Copied to clipboard
Yuya Sawada, Hiroki Ouchi, Yuichiro Yasui, Hiroki Teranishi, Yuji Matsumoto, Taro Watanabe, Masayuki Ishii
| Challenge: | Existing information extraction systems are not able to accurately capture organizational changes. |
| Approach: | They propose a task to extract corporate history events related to organizational changes by identifying company names before and after each event, as well as the corresponding date. |
| Outcome: | The proposed task is designed to identify company names before and after an event, as well as the corresponding date. |
Global Entity Disambiguation with BERT (2022.naacl-main)
Copied to clipboard
| Challenge: | Entity disambiguation (ED) is a task of assigning mentions to referent entities in a knowledge base. |
| Approach: | They propose a global entity disambiguation (ED) model based on BERT . they train the model using a large entity-annotated corpus obtained from Wikipedia . |
| Outcome: | The proposed model can disambiguate masked entities based on words and non-masked ones at the inference time. |
Entity Profile Generation and Reasoning with LLMs for Entity Alignment (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Entity alignment is a process of identifying and linking equivalent entities across knowledge graphs . only a small fraction of these entities are aligned . |
| Approach: | They propose a method that combines large language models with entity embeddings to align entities. |
| Outcome: | ProLEA is a method that combines large language models with entity embeddings to improve alignment accuracy, robustness, and explainability. |
Construction of Large-scale English Verbal Multiword Expression Annotated Corpus (L18-1)
Copied to clipboard
| Challenge: | In this paper, we focus on verbal MWEs, whose accurate recognition is challenging because they could be discontinuous. |
| Approach: | They conduct large-scale annotations of VMWEs on the Wall Street Journal portion of Ontonotes . they first construct a VMwe dictionary based on the english-language Wiktionary . |
| Outcome: | The proposed resource annotates 7,833 VMWE instances belonging to various categories . the authors hope the results will help to develop models for MWE recognition and dependency parsing . |
PDFAnno: a Web-based Linguistic Annotation Tool for PDF Documents (L18-1)
Copied to clipboard
| Challenge: | Currently, linguistic annotation tools for PDF documents focus on plain-text documents. |
| Approach: | They propose a web-based linguistic annotation tool for PDF documents . it offers functions for various types of linguistic annotations directly on PDF . |
| Outcome: | The proposed tool can annotate on PDF documents with named entity, dependency relation, and coreference chain. |
A Dataset for Pharmacovigilance in German, French, and Japanese: Annotating Adverse Drug Reactions across Languages (2024.lrec-main)
Copied to clipboard
Lisa Raithel, Hui-Syuan Yeh, Shuntaro Yada, Cyril Grouin, Thomas Lavergne, Aurélie Névéol, Patrick Paroubek, Philippe Thomas, Tomohiro Nishiyama, Sebastian Möller, Eiji Aramaki, Yuji Matsumoto, Roland Roller, Pierre Zweigenbaum
| Challenge: | Existing clinical corpora mostly revolves around scientific articles in English . existing literature is limited to only a few scientific articles . |
| Approach: | They propose to use user-generated data sources to uncover adverse drug reactions . existing clinical corpora mostly revolves around scientific articles in english . authors provide statistics to highlight certain challenges associated with the corpus . |
| Outcome: | The proposed corpus includes 12 entity types, four attribute types, and 13 relation types . it provides strong baselines for extracting entities and relations between entities . |
Sudachi: a Japanese Tokenizer for Business (L18-1)
Copied to clipboard
| Challenge: | Lack of token unit compatibility is one of the critical problems of Japanese language resources. |
| Approach: | They develop a Japanese tokenizer called Sudachi and its accompanying dictionary . they use multi-granular output and normalization of notation variations to improve tokenization . |
| Outcome: | The proposed tokenizer and dictionary improve tokenization in Japanese for business use. |
Post Persona Alignment for Multi-Session Dialogue Generation (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods for multi-session persona-based dialogue generation typically retrieve persona information before response generation, which can constrain diversity and result in generic outputs. |
| Approach: | They propose a two-stage framework that reverses the process of retrieving persona information before response generation. |
| Outcome: | Experiments on multi-session persona-based dialogue data show that the proposed framework outperforms existing methods in consistency, diversity, and persona relevance. |
Coordination Boundary Identification without Labeled Data for Compound Terms Disambiguation (2020.coling-main)
Copied to clipboard
Yuya Sawada, Takashi Wada, Takayoshi Shibahara, Hiroki Teranishi, Shuhei Kondo, Hiroyuki Shindo, Taro Watanabe, Yuji Matsumoto
| Challenge: | a new method for nominal coordination boundary identification is proposed . it uses pre-trained word embeddings to measure similarities of words and detects the span of coordination . |
| Approach: | They propose a method for nominal coordination boundary identification that uses pre-trained word embeddings to measure similarities of words and detects the span of coordination. |
| Outcome: | The proposed method can identify coordination boundaries without training on labeled data . it is comparable to a recent supervised method for the case when the coordinator conjoins simple noun phrases. |