Koichi Takeda

• Ph. D in Informatics
• Managing Director at Nagoya University

Driving video captioning aims to automatically generate descriptions for videos from driving recorders. Driving video captions are generally required to describe first-person driving behaviors which implicitly characterize the driving videos but are challenging to anchor to concrete visual evidence. To generate captions with better driving behavior...

In the Autonomous Driving (AD) scenario, accurate, informative, and understandable descriptions of the traffic conditions and the ego-vehicle motions can increase the interpretability of an autonomous driving system to the vehicle user. End-to-end free-form video captioning is a straightforward vision-to-text task to address such needs. However, in...

When a sports match is broadcast, X users often enjoy sharing the comment and it is possible to roughly understand a match’s progress by reading these posts. However, because of the diverse nature of posts, it can be challenging to quickly grasp a match’s progress. In this study, we focus on soccer matches and work on building a system to generate...

Prediction of the future citation counts of papers is increasingly important to find interesting papers among an ever-growing number of papers. Although a paper's main text is an important factor for citation count prediction, it is difficult to handle in machine learning models because the main text is typically very long; thus previous studies ha...

Citation count prediction is the task of predicting the future citation counts of academic papers, which is particularly useful for estimating the future impacts of an ever-growing number of academic papers. Although there have been many studies on citation count prediction, they are not applicable to predicting the citation counts of newly publish...

In recent years, neural machine translation (NMT) has been widely used in everyday life. However, the current NMT lacks a mechanism to adjust the difficulty level of translations to match the user's language level. Additionally, due to the bias in the training data for NMT, translations of simple source sentences are often produced with complex wor...

Large language models (LLMs) are supposed to acquire unconscious human knowledge and feelings, such as social common sense and biases, by training models from large amounts of text. However, it is not clear how much the sentiments of specific social groups can be captured in various LLMs. In this study, we focus on social groups defined in terms of...

In practice, even a well-trained neural machine translation (NMT) model can still make biased inferences on the training set due to distribution shifts. For the human learning process, if we can not reproduce something correctly after learning it multiple times, we consider it to be more difficult. Likewise, a training example causing a large discr...

The semantic frame induction tasks are defined as a clustering of words into the frames that they evoke, and a clustering of their arguments according to the frame element roles that they should fill. In this paper, we address the latter task of argument clustering, which aims to acquire frame element knowledge, and propose a method that applies de...

Recent progress in sentence embedding, which represents the meaning of a sentence as a point in a vector space, has achieved high performance on tasks such as a semantic textual similarity (STS) task. However, sentence representations as a point in a vector space can express only a part of the diverse information that sentences have, such as asymme...

Recent studies have demonstrated the usefulness of contextualized word embeddings in unsupervised semantic frame induction. However, they have also revealed that generic contextualized embeddings are not always consistent with human intuitions about semantic frames, which causes unsatisfactory performance for frame induction based on contextualized...

We have recently seen many successful applications of sentence embedding methods. It has not been well understood, however, what kind of properties are captured in the resulting sentence embeddings, depending on the supervision signals. In this paper, we focus on two types of sentence embeddings obtained by using natural language inference (NLI) da...

This paper explores a variant of automatic headline generation methods, where a generated headline is required to include a given phrase such as a company or a product name. Previous methods using Transformer-based models generate a headline including a given phrase by providing the encoder with additional information corresponding to the given phr...

Contextualized word representations have proven useful for various natural language processing tasks. However, it remains unclear to what extent these representations can cover hand-coded semantic information such as semantic frames, which specify the semantic role of the arguments associated with a predicate. In this paper, we focus on verbs that...

Recent studies on semantic frame induction show that relatively high performance has been achieved by using clustering-based methods with contextualized word embeddings. However, there are two potential drawbacks to these methods: one is that they focus too much on the superficial information of the frame-evoking verb and the other is that they ten...

Sentence embedding methods using natural language inference (NLI) datasets have been successfully applied to various tasks. However, these methods are only available for limited languages due to relying heavily on the large NLI datasets. In this paper, we propose DefSent, a sentence embedding method that uses definition sentences from a word dictio...

In the field of machine learning, the well-trained model is assumed to be able to recover the training labels, i.e. the synthetic labels predicted by the model should be as close to the ground-truth labels as possible. Inspired by this, we propose a self-guided curriculum strategy to encourage the learning of neural machine translation (NMT) models...

This paper describes our submission of the WMT 2020 Shared Task on Sentence Level Direct Assessment, Quality Estimation (QE). In this study, we empirically reveal the mismatching issue when directly adopting BERTScore (Zhang et al., 2020) to QE. Specifically, there exist lots of mismatching errors between the source sentence and translated candidat...

This paper describes our submission of the WMT 2020 Shared Task on Sentence Level Direct Assessment, Quality Estimation (QE). In this study, we empirically reveal the \textit{mismatching issue} when directly adopting BERTScore to QE. Specifically, there exist lots of mismatching errors between the source sentence and translated candidate sentence w...

It has been reported that a person’s remarks and behaviors reflect the person’s personality. Several recent studies have shown that textual information of user posts and user behaviors such as liking and reblogging the specific posts are useful for predicting the personality of Social Networking Service (SNS) users. However, less attention has been...

The present article addresses an attempt to apply questions in university entrance examinations to the evaluation of textual entailment recognition. Questions in several fields, such as history and politics, primarily test the examinee’s knowledge ...

This paper introduces an overview of the RITE (Recognizing Inference in TExt) task in NTCIR-9. We evaluate systems that automatically recognize entailment, paraphrase, and contradiction between two texts written in Japanese, Simplified Chinese, or Traditional Chinese. The task consists of four subtasks: Binary classification of entailment (BC); Mul...

There are increasingly visible demands for structured/ unstructured information integration and advanced analytics. However, conventional database technology has not been able to present a robust and practical implementation of a truly integrated architecture for such purposes. After working on several industrial applications (in particular, in the...

This paper describes MedTAKMI-CDI, an online analytical processing system that enables the interactive discovery of knowledge for clinical decision intelligence (CDI). CDI supports decision making by providing in-depth analysis of clinical data from multiple sources. We discuss the fundamental challenges we faced and explain how we met those challe...

This paper describes the application of IBM TAKMI® for Biomedical Documents to facilitate knowledge discovery from the very large text databases characteristic of life science and healthcare applications. This set of tools, designated MedTAKMI, is an extension of the TAKMI (Text Analysis and Knowledge MIning) system originally developed for text mi...

In the Patent Retrieval Task in NTCIR-4 Workshop, the search topic is the claim in a patent document, so we use the claim text and the IPC information for the similarity calculations between the search topic and each patent document in the collection. We examined the ef- fectiveness of the similarity measure be- tween IPCs and the term weighting fo...

This paper describes methods for relating (threading) multiple newspaper articles, and for visualizing various characteristics of them by using a directed graph. A set of articles is represented by a set of word vectors, and the similarity between the vectors is then calculated. The graph is constructed from the similarity matrix. By applying some...

This paper proposes the use of "patternbased " context-free grammars as a basis for building machine translation (MT) sys- tems, which are now being adopted as personal tools by a broad range of users in the cyberspace society. We discuss major requirements for such tools, including easy customization for diverse domains, the efficiency of the tran...

In this paper, we describe a machine translation system called PalmTree which uses the "patternbased " approach as a fundamental framework. The pure pattern-based translation framework has several issues. One is the performance due to using many rules in the parsing stage, and the other is inefficiency of usage of translation patterns due to the ex...

Machine tra.nslation (MT) has recently been formula. ted in terms of constra.int-based knowledge representa.tion a.nd unifica.tion theories but it is becoming more and more evident tha.t it is not possible to design a. practica.l MT system without a.n a.dequa.te method of ha.ndling mismatches between sema.ntic representations in the source and targ...

this paper, we introduce two such projects -- text mining and site outlining -- conducted at the Tokyo Research Laboratory, IBM Research

In this paper, we propose a novel approach for personalized Web Knowledge management by incorporating dynamic (e.g., news article) and static (e.g.,technical papers) Web contents. We can integrate these two types of information by using intelligent crawling and meta-data creation and provide a visual interface to personalize and navigate through th...

In this paper we review studies of the growth of the Internet and technologies that are useful for information search and retrieval on the Web. We present data on the Internet from several different sources, e.g., current as well as projected number of users, hosts, and Web sites. Although numerical figures vary, overall trends cited by the sources...

We propose a novel visual interface for digital libraries, which employs a collection of views and a compositional mechanism of views. While the views capture and visualize specific aspects of digital libraries, their compositions can be utilized to give deeper and richer insights into a set of retrieved data. Two types of compositions-superimposit...

In this paper, we describe a "pattern-based" machine translation (MT) approach that we followed in designing a personal tool for users who have access to large volumes of text in languages other than their own, such as WWW pages. Some of the critical issues involved in the design of such a tool include easy customization for diverse domains, the ef...

In this paper, we describe the acquisition and organization of knowledge sources for machine translation (MT) systems. It has been pointed out by many users that one of the most annoying things about MT systems is the repeated occurrence of identical errors in word sense and attachment disambignuation. We show the limitations of a conventional user...

In this paper, we describe an implementation of object-oriented knowledge sources and functions for knowledge-based machine translation, where the meanings of sentences are represented by conceptual "objects." While parsing and generation are viewed as map-pings between syntactic and conceptual representations, such functions as paraphrasing, abstr...

Shalt2 is a knowledge-based machine translation system with a symmetric architecture. The grammar rules, mapping rules between syntactic and conceptual (semantic) representations, and transfer rules for conceptual paraphrasing are all bi-directional knowledge sources used by both a parser and a generator.

Many models of the processing of printed or spoken words or objects or faces propose that systems of local representations of the forms of such stimuli—lexicons—exist. This is denied by partisans of the distributed‐representation connectionist approach to cognitive modelling. An experimental paradigm of key theoretical importance here is lexical de...

This paper describes the analysis and generation grammars for English and Japanese as they were employed in the KBMT-89 program. We discuss word order, coordination, subcategorization, morphological rules, rule ordering and bi-directional grammars.

This paper describes an experimental expert system for proofreading Japanese text. The system is called CRITAC (CRITiquing using ACcumulated knowledge). It can detect typographical errors, Kana-to-Kanji conversion errors, and stylistic errors in Japanese text. We describe the basic concepts and features of CRITAC, including preprocessing of text, a...

In this paper, we consider two fundamental properties of nested relations: minimum and uniqueness. By nested relations we mean relations over simple and/or set-valued domains. It is shown that the complexity of deciding whether or not a given first normal form relation has a nested form with no more than K tuples is NP-complete. A template dependen...

CRITAC (CRITiquing using A Ccumulated knowledge) is an experimental expert system for proofreading Japenese text. It detects mistypes, Kana-to-Kanji misconversions, and stylistic errors. This system combines Prolog-coded heuristic knowledge with conventional Japanese text processing techniques which involve heavy computation and access to large lan...

There are two methods to handle redundancies caused by dependency constraints in relational databases. The first method, decomposition into relations in high-level normal form, has been studied by various authors, since it is very important for logical database design. The second method, representation by unnormalized form, is discussed in this pap...

In this paper we review studies on the growth of the Internet and technologies which are useful for information search and retrieval on the Web. In the rst section, we present data on the Internet from several dierent sources, e.g., current as well as projected number of users, hosts and Web sites. Although the numerical gures vary, the overall tre...

In this paper we discuss sentence generation strategy for pattern-based machine translation and their computational properties. Even though sentence generation in general is known to be computationally expensive, there exists a polynomial-time algorithm for synchronized sentence analysis/generation for "fixed" pattern-based (and its variant) gramma...

paper presents an overview of the ACLIA (Advanced Cross-Lingual Information Access) task cluster at NTCIR-8. The task overview includes: a definition of and motivation for the evaluation; a description of the complex and factoid question types evaluated; the document sources and exchange formats selected and/or defined; the official metrics used in...

Shall2 is a knowledge-based machine translation sys- tem with a symmetric architecture. The grammar rules, mapping rules between syntactic and conceptual (semantic) representations, and transfer rules for con- ceptual paraphrasing are all bi-directional knowledge sources used by both a parser and a generator.

Kyoto University (京都大学) 0048 新制・課程博士 博士(情報学) 甲第15521号 情博第379号 新制/情/70 27999 2010-03-23 京都大学大学院情報学研究科知能情報学専攻 (主査)教授 黒橋 禎夫, 教授 西田 豊明, 教授 河原 達也 学位規則第4条第1項該当

http://www.tulips.tsukuba.ac.jp/mylimedio/dl/page.do?issueid=1000078&tocid=100085297&page=79-86