Searcharxiv⌕ Search

arXiv subjects

Jun Kato

Publications and source records attributed to Jun Kato.

12 recordsLinked to original sources
arXiv2025

Inter(sectional) Alia(s): Ambiguity in Voice Agent Identity via Intersectional Japanese Self-Referents

Conversational agents that mimic people have raised questions about the ethics of anthropomorphizing machines with human social identity cues. Critics have also questioned assumptions of identity neutrality in humanlike agents. Recent work has revealed that intersectional Japanese pronouns can elicit complex and sometimes evasive impressions of agent identity. Yet, the role of other "neutral" non-pronominal self-referents (NPSR) and voice as a socially expressive medium remains unexplored. In a crowdsourcing study, Japanese participants (N = 204) evaluated three ChatGPT voices (Juniper, Breeze, and Ember) using seven self-referents. We found strong evidence of voice gendering alongside the potential of intersectional self-referents to evade gendering, i.e., ambiguity through neutrality and elusiveness. Notably, perceptions of age and formality intersected with gendering as per sociolinguistic theories, especially boku and watakushi. This work provides a nuanced take on agent identity perceptions and champions intersectional and culturally-sensitive work on voice agents.

cs.HC↗
arXiv2025

Super Kawaii Vocalics: Amplifying the "Cute" Factor in Computer Voice

"Kawaii" is the Japanese concept of cute, which carries sociocultural connotations related to social identities and emotional responses. Yet, virtually all work to date has focused on the visual side of kawaii, including in studies of computer agents and social robots. In pursuit of formalizing the new science of kawaii vocalics, we explored what elements of voice relate to kawaii and how they might be manipulated, manually and automatically. We conducted a four-phase study (grand N = 512) with two varieties of computer voices: text-to-speech (TTS) and game character voices. We found kawaii "sweet spots" through manipulation of fundamental and formant frequencies, but only for certain voices and to a certain extent. Findings also suggest a ceiling effect for the kawaii vocalics of certain voices. We offer empirical validation of the preliminary kawaii vocalics model and an elementary method for manipulating kawaii perceptions of computer voice.

cs.HC↗
arXiv2024

A Practical Style Transfer Pipeline for 3D Animation: Insights from Production R&D

Our animation studio has developed a practical style transfer pipeline for creating stylized 3D animation, which is suitable for complex real-world production. This paper presents the insights from our development process, where we explored various options to balance quality, artist control, and workload, leading to several key decisions. For example, we chose patch-based texture synthesis over machine learning for better control and to avoid training data issues. We also addressed specifying style exemplars, managing multiple colors within a scene, controlling outlines and shadows, and reducing temporal noise. These insights were used to further refine our pipeline, ultimately enabling us to produce an experimental short film showcasing various styles.

cs.GR↗
arXiv2024

Deep Graph Attention Networks

Graphs are useful for representing various realworld objects. However, graph neural networks (GNNs) tend to suffer from over-smoothing, where the representations of nodes of different classes become similar as the number of layers increases, leading to performance degradation. A method that does not require protracted tuning of the number of layers is needed to effectively construct a graph attention network (GAT), a type of GNN. Therefore, we introduce a method called "DeepGAT" for predicting the class to which nodes belong in a deep GAT. It avoids over-smoothing in a GAT by ensuring that nodes in different classes are not similar at each layer. Using DeepGAT to predict class labels, a 15-layer network is constructed without the need to tune the number of layers. DeepGAT prevented over-smoothing and achieved a 15-layer GAT with similar performance to a 2-layer GAT, as indicated by the similar attention coefficients. DeepGAT enables the training of a large network to acquire similar attention coefficients to a network with few layers. It avoids the over-smoothing problem and obviates the need to tune the number of layers, thus saving time and enhancing GNN performance.

cs.LG↗
arXiv2024

Small energy scattering for radial solutions to the generalized Zakharov system

We prove the small energy scattering for the three-dimensional generalized Zakharov system with radial symmetry based on the idea by Guo and Nakanishi (2014), which treats the usual Zakharov system. For the proof, we use the frequency-localized normal form reduction, and the radially improved Strichartz estimates. The relation between the solution to the integral equations, which includes the unusual boundary terms, and the original differential equations is also considered.

math.AP↗
arXiv2021

Guided Optimization for Image Processing Pipelines

Writing high-performance image processing code is challenging and labor-intensive. The Halide programming language simplifies this task by decoupling high-level algorithms from "schedules" which optimize their implementation. However, even with this abstraction, it is still challenging for Halide programmers to understand complicated scheduling strategies and productively write valid, optimized schedules. To address this, we propose a programming support method called "guided optimization." Guided optimization provides programmers a set of valid optimization options and interactive feedback about their current choices, which enables them to comprehend and efficiently optimize image processing code without the time-consuming trial-and-error process of traditional text editors. We implemented a proof-of-concept system, Roly-poly, which integrates guided optimization, program visualization, and schedule cost estimation to support the comprehension and development of efficient Halide image processing code. We conducted a user study with novice Halide programmers and confirmed that Roly-poly and its guided optimization was informative, increased productivity, and resulted in higher-performing schedules in less time.

cs.HC↗
arXiv2020

Lyric Video Analysis Using Text Detection and Tracking

We attempt to recognize and track lyric words in lyric videos. Lyric video is a music video showing the lyric words of a song. The main characteristic of lyric videos is that the lyric words are shown at frames synchronously with the music. The difficulty of recognizing and tracking the lyric words is that (1) the words are often decorated and geometrically distorted and (2) the words move arbitrarily and drastically in the video frame. The purpose of this paper is to analyze the motion of the lyric words in lyric videos, as the first step of automatic lyric video generation. In order to analyze the motion of lyric words, we first apply a state-of-the-art scene text detector and recognizer to each video frame. Then, lyric-frame matching is performed to establish the optimal correspondence between lyric words and the frames. After fixing the motion trajectories of individual lyric words from correspondence, we analyze the trajectories of the lyric words by k-medoids clustering and dynamic time warping (DTW).

cs.CV↗
arXiv2020

ClassCode: An Interactive Teaching and Learning Environment for Programming Education in Classrooms

Programming education is becoming important as demands on computer literacy and coding skills are growing. Despite the increasing popularity of interactive online learning systems, many programming courses in schools have not changed their teaching format from the conventional classroom setting. We see two research opportunities here. Students may have diverse expertise and experience in programming. Thus, particular content and teaching speed can be disengaging for experienced students or discouraging for novice learners. In a large classroom, instructors cannot oversee the learning progress of each student, and have difficulty matching teaching materials with the comprehension level of individual students. We present ClassCode, a web-based environment tailored to programming education in classrooms. Students can take online tutorials prepared by instructors at their own pace. They can then deepen their understandings by performing interactive coding exercises interleaved within tutorials. ClassCode tracks all interactions by each student, and summarizes them to instructors. This serves as a progress report, facilitating the instructors to provide additional explanations in-situ or revise course materials. Our user evaluation through a small lecture and expert review by instructors and teaching assistants confirm the potential of ClassCode by uncovering how it could address issues in existing programming courses at universities.

cs.CY↗
arXiv2016

Making Bipartite Graphs DM-irreducible

The Dulmage--Mendelsohn decomposition (or the DM-decomposition) gives a unique partition of the vertex set of a bipartite graph reflecting the structure of all the maximum matchings therein. A bipartite graph is said to be DM-irreducible if its DM-decomposition consists of a single component. In this paper, we focus on the problem of making a given bipartite graph DM-irreducible by adding edges. When the input bipartite graph is balanced (i.e., both sides have the same number of vertices) and has a perfect matching, this problem is equivalent to making a directed graph strongly connected by adding edges, for which the minimum number of additional edges was characterized by Eswaran and Tarjan (1976). We give a general solution to this problem, which is divided into three parts. We first show that our problem can be formulated as a special case of a general framework of covering supermodular functions, which was introduced by Frank and Jord\'an (1995) to investigate the directed connectivity augmentation problem. Secondly, when the input graph is not balanced, the problem is solved via matroid intersection. This result can be extended to the minimum cost version in which the addition of an edge gives rise to an individual cost. Thirdly, for balanced input graphs, we devise a combinatorial algorithm that finds a minimum number of additional edges to attain the DM-irreducibility, while the minimum cost version of this problem is NP-hard. These results also lead to min-max characterizations of the minimum number, which generalize the result of Eswaran and Tarjan.

cs.DM↗
arXiv2015

A remark on a priori estimate for the Navier-Stokes equations with the Coriolis force

The Cauchy problem for the Navier-Stokes equations with the Coriolis force is considered. It is proved that a similar a priori estimate, which is derived for the Navier-Stokes equations by Lei and Lin [11], holds under the effect of the Coriolis force. As an application existence of a unique global solution for arbitrary speed of rotation is proved, as well as its asymptotic behavior.

math.AP↗
arXiv2010

A new proof of long range scattering for critical nonlinear Schrödinger equations

We present a new proof of global existence and long range scattering, from small initial data, for the one-dimensional cubic gauge invariant nonlinear Schrödinger equation, and for Hartree equations in dimension $n \geq 2$. The proof relies on an analysis in Fourier space, related to the recent works of Germain, Masmoudi and Shatah on space-time resonances. An interesting feature of our approach is that we are able to identify the long range phase correction term through a very natural stationary phase argument.

math.AP↗