arXiv subjects

Hiroyuki Sato

Publications and source records attributed to Hiroyuki Sato.

At least 19 recordsLinked to original sources

arXiv2026

GPU-Native Real-Time Textured Mesh Reconstruction with Seam-Aware Multi-View Radiometric Correction

We present a GPU-native volumetric-video system for real-time reconstruction of graphics-compatible textured triangle meshes from multiple RGB-D sensors. Existing TSDF systems accelerate fusion and mesh extraction, but extending them to per-frame textured meshing adds CPU mesh processing, CPU-side texturing work, or extra CPU-GPU transfers between stages. After input staging, our method keeps the reconstruction and texturing stages (TSDF fusion, Marching Cubes, connected-component noise removal, view-based texture mapping, inpainting, and radiometric seam correction) resident on the GPU; no intermediate mesh or image data returns to the host between these stages. A fixed-size-array representation of mesh topology gives constant-time neighbor access on the GPU and makes the downstream mesh-level operations data-parallel. Because we texture each face from a single camera, exposure seams that vary smoothly across each image remain; we estimate a per-camera low-frequency gain field and bake it into the atlas on the GPU. On the CMU Panoptic dataset, our method is faster than the baselines at every tested voxel resolution and number of sensors, achieves peak speedups of more than an order of magnitude over CUDA-accelerated nvblox and CPU-based Open3D, and approaches an offline textured-meshing baseline in perceptual re-rendering quality.

cs.CV↗

arXiv2026

Conversational Capture: A Trajectory-Level Framework for Evaluating Generative Engine Optimization in Multi-turn Human-Agent Interaction

Generative Engine Optimization (GEO) shapes content to increase its likelihood of being cited by answer engines built on retrieval-augmented large language models. GEO is typically evaluated as a single-turn property: for a fixed query, an evaluator measures a source's visibility in one answer. We argue that the single answer is an inadequate unit of analysis. Human-agent information seeking forms a closed loop: the agent's answer changes the user's beliefs and therefore the next question, which in turn determines what the agent retrieves. We introduce conversational capture, a phenomenon in which a source cited early becomes substantially more likely to be cited again. Capture operates through a machine-side channel, history-conditioned retrieval, and a human-side channel, follow-up questions directed toward the captured source. We formalize the interaction as a two-layer closed-loop system and derive trajectory-level constructs: cumulative conversational visibility; a direct/feedback decomposition of trajectory gain; a nested split of the feedback term into machine-side and human-side channels; a capture coefficient; a compounding ratio; and a misranking diagnostic. Using reinforcement-process (Pólya-urn) theory, we prove that the feedback term is zero under single-turn evaluation and that GEO's cumulative payoff grows superlinearly with conversation length while capture develops. A model-derived illustration shows that the feedback term can exceed the direct term, the compounding ratio exceeds two within ten turns, and single-turn and trajectory rankings agree only weakly (Kendall's $τ= 0.4$). We connect the human channel to information foraging, trust calibration, and Bayesian persuasion, and discuss design implications for answer engines.

cs.HC↗

arXiv2026

Impacts of Single-objective Landscapes on Multi-objective Optimization

This work revealed a relationship between a multi-objective optimization problem and single-objective optimization problems that exist in the multi-objective problem. This work focused on combinatorial problems and investigated the relations between the local optima networks of the single-objective problems and the Pareto optima network of the multi-objective problem. Each of their networks has a graph structure. We divided the entire network into subgraphs. Each subgraph was called a component and characterized by overlapping relations between the single-objective local optima networks and the multi-objective Pareto optima network. Results on multi-objective landscape problems showed that most Pareto optimal solutions were reachable from the single-objective local optimal solutions. This tendency was emphasized by increasing the number of objectives and the objective correlation. The number of co-variables impacted the number of cross-link relations between the single-objective local optima networks and the multi-objective Pareto optima network. The results suggested that searching for single-objective problems is a clue to multi-objective optimization.

cs.NE↗

arXiv2026

Second-order geometry and Riemannian Newton-type methods for optimization on the indefinite Stiefel manifold

This paper investigates the second-order geometry of the indefinite Stiefel manifold and derives explicit formulas for the Levi-Civita connection and the Riemannian Hessian under two generalized canonical metrics. We discuss Riemannian Newton's method, in which Newton's equation is solved by the linear conjugate gradient method in a fixed tangent space, and the Riemannian trust-region method with the truncated conjugate gradient method. Numerical experiments for trace minimization problems demonstrate the robustness of the trust-region method over several problem sizes and in near-singular settings where eigenvalues of the constraint matrix approach zero.

math.OC↗

arXiv2026

Riemannian optimization framework on the generalized quaternionic Stiefel Manifold

This paper introduces the generalized quaternionic Stiefel manifold and studies its geometry for Riemannian optimization. We clarify its relationships with existing manifolds, especially the real generalized Stiefel manifold and the quaternionic Stiefel manifold, and derive explicit expressions for several geometric quantities on the proposed manifold. In particular, the generalized quaternionic Stiefel manifold is regarded as a real Riemannian manifold, and expressions for the tangent space, normal space, the orthogonal projection onto the tangent space, a retraction, and a vector transport on this manifold are derived. Some numerical experiments for the generalized quaternionic eigenvalue problem and a version of quaternionic canonical correlation analysis are performed by Riemannian optimization methods to demonstrate the viability of the proposed optimization framework.

math.OC↗

arXiv2026

Structural Feature Engineering for Generative Engine Optimization: How Content Structure Shapes Citation Behavior

The proliferation of AI-powered search engines has shifted information discovery from traditional link-based retrieval to direct answer generation with selective source citation, creating new challenges for content visibility. While existing Generative Engine Optimization (GEO) approaches focus primarily on semantic content modification, the role of structural features in influencing citation behavior remains underexplored. In this paper, we propose GEO-SFE, a systematic framework for structural feature engineering in generative engine optimization. Our approach decomposes content structure into three hierarchical levels: macro-structure (document architecture), meso-structure (information chunking), and micro-structure (visual emphasis), and models their impact on citation probability across different generative engine architectures. We develop architecture-aware optimization strategies and predictive models that preserve semantic integrity while improving structural effectiveness. Experimental evaluation across six mainstream generative engines demonstrates consistent improvements in citation rate (17.3 percent) and subjective quality (18.5 percent), validating the effectiveness and generalizability of the proposed framework. This work establishes structural optimization as a foundational component of GEO, providing a data-driven methodology for enhancing content visibility in LLM-powered information ecosystems.

cs.CL↗

arXiv2025

Decentralized Multi-Agent System with Trust-Aware Communication

The emergence of Large Language Models (LLMs) is rapidly accelerating the development of autonomous multi-agent systems (MAS), paving the way for the Internet of Agents. However, traditional centralized MAS architectures present significant challenges, including single points of failure, vulnerability to censorship, inherent scalability limitations, and critical trust issues. We propose a novel Decentralized Multi-Agent System (DMAS) architecture designed to overcome these fundamental problems by enabling trust-aware, scalable, and censorship-resistant interactions among autonomous agents. Our DMAS features a decentralized agent runtime underpinned by a blockchain-based architecture. We formalize a trust-aware communication protocol that leverages cryptographic primitives and on-chain operations to provide security properties: verifiable interaction cycles, communication integrity, authenticity, non-repudiation, and conditional confidentiality, which we further substantiate through a comprehensive security analysis. Our performance analysis validates the DMAS as a scalable and efficient solution for building trustworthy multi-agent systems.

cs.MA↗

arXiv2025

DynTaskMAS: A Dynamic Task Graph-driven Framework for Asynchronous and Parallel LLM-based Multi-Agent Systems

The emergence of Large Language Models (LLMs) in Multi-Agent Systems (MAS) has opened new possibilities for artificial intelligence, yet current implementations face significant challenges in resource management, task coordination, and system efficiency. While existing frameworks demonstrate the potential of LLM-based agents in collaborative problem-solving, they often lack sophisticated mechanisms for parallel execution and dynamic task management. This paper introduces DynTaskMAS, a novel framework that orchestrates asynchronous and parallel operations in LLM-based MAS through dynamic task graphs. The framework features four key innovations: (1) a Dynamic Task Graph Generator that intelligently decomposes complex tasks while maintaining logical dependencies, (2) an Asynchronous Parallel Execution Engine that optimizes resource utilization through efficient task scheduling, (3) a Semantic-Aware Context Management System that enables efficient information sharing among agents, and (4) an Adaptive Workflow Manager that dynamically optimizes system performance. Experimental evaluations demonstrate that DynTaskMAS achieves significant improvements over traditional approaches: a 21-33% reduction in execution time across task complexities (with higher gains for more complex tasks), a 35.4% improvement in resource utilization (from 65% to 88%), and near-linear throughput scaling up to 16 concurrent agents (3.47X improvement for 4X agents). Our framework establishes a foundation for building scalable, high-performance LLM-based multi-agent systems capable of handling complex, dynamic tasks efficiently.

cs.MA↗

arXiv2025

Modified Armijo line search in optimization on Riemannian submanifolds with reduced computational cost

For optimization problems on Riemannian manifolds, many types of globally convergent algorithms have been proposed, and they are often equipped with the Riemannian version of the Armijo line search for global convergence. Such existing methods need to compute the value of a retraction mapping regarding the search direction several times at each iteration; this may result in high computational costs, particularly if computing the value of the retraction is expensive. To address this issue, this study focuses on Riemannian submanifolds of the Euclidean spaces and proposes a novel Riemannian line search that achieves lower computational cost by incorporating a new strategy that computes the retraction only when inevitable. A class of Riemannian optimization algorithms, including the steepest descent and Newton methods, with the new line search strategy is proposed and proved to be globally convergent. Furthermore, numerical experiments on solving optimization problems on several types of Riemannian submanifolds illustrate that the proposed methods are superior to the standard Riemannian Armijo line search-based methods.

math.OC↗

arXiv2024

BUPD: A Bayesian under-parameterized basket design with the unit information prior in oncology trials

Basket trials in oncology enroll multiple patients with cancer harboring identical gene alterations and evaluate their response to targeted therapies across cancer types. Several existing methods have extended a Bayesian hierarchical model borrowing information on the response rates in different cancer types to account for the heterogeneity of drug effects. However, these methods rely on several pre-specified parameters to account for the heterogeneity of response rates among different cancer types. Here, we propose a novel Bayesian under-parameterized basket design with a unit information prior (BUPD) that uses only one (or two) pre-specified parameters to control the amount of information borrowed among cancer types, considering the heterogeneity of response rates. BUPD adapts the unit information prior approach, originally developed for borrowing information from historical clinical trial data, to enable mutual information borrowing between two cancer types. BUPD enables flexible controls of the type 1 error rate and power by explicitly specifying the strength of borrowing while providing interpretable estimations of response rates. Simulation studies revealed that BUPD reduced the type 1 error rate in scenarios with few ineffective cancer types and improved the power in scenarios with few effective cancer types better than five existing methods. This study also illustrated the efficiency of BUPD using response rates from a real basket trial.

stat.ME↗

arXiv2024

1-D CNN-Based Online Signature Verification with Federated Learning

Online signature verification plays a pivotal role in security infrastructures. However, conventional online signature verification models pose significant risks to data privacy, especially during training processes. To mitigate these concerns, we propose a novel federated learning framework that leverages 1-D Convolutional Neural Networks (CNN) for online signature verification. Furthermore, our experiments demonstrate the effectiveness of our framework regarding 1-D CNN and federated learning. Particularly, the experiment results highlight that our framework 1) minimizes local computational resources; 2) enhances transfer effects with substantial initialization data; 3) presents remarkable scalability. The centralized 1-D CNN model achieves an Equal Error Rate (EER) of 3.33% and an accuracy of 96.25%. Meanwhile, configurations with 2, 5, and 10 agents yield EERs of 5.42%, 5.83%, and 5.63%, along with accuracies of 95.21%, 94.17%, and 94.06%, respectively.

cs.CR↗

arXiv2024

Model-Driven Security Analysis of Self-Sovereign Identity Systems

Best practices of self-sovereign identity (SSI) are being intensively explored in academia and industry. Reusable solutions obtained from best practices are generalized as architectural patterns for systematic analysis and design reference, which significantly boosts productivity and increases the dependability of future implementations. For security-sensitive projects, architects make architectural decisions with careful consideration of security issues and solutions based on formal analysis and experiment results. In this paper, we propose a model-driven security analysis framework for analyzing architectural patterns of SSI systems with respect to a threat model built on our investigation of real-world security concerns. Our framework mechanizes a modeling language to formalize patterns and threats with security properties in temporal logic and automatically generates programs for verification via model checking. Besides, we present typical vulnerable patterns verified by SecureSSI, a standalone integrated development environment, integrating commonly used pattern and attacker models to practicalize our framework.

cs.CR↗

arXiv2024

BlockFusion: Expandable 3D Scene Generation using Latent Tri-plane Extrapolation

We present BlockFusion, a diffusion-based model that generates 3D scenes as unit blocks and seamlessly incorporates new blocks to extend the scene. BlockFusion is trained using datasets of 3D blocks that are randomly cropped from complete 3D scene meshes. Through per-block fitting, all training blocks are converted into the hybrid neural fields: with a tri-plane containing the geometry features, followed by a Multi-layer Perceptron (MLP) for decoding the signed distance values. A variational auto-encoder is employed to compress the tri-planes into the latent tri-plane space, on which the denoising diffusion process is performed. Diffusion applied to the latent representations allows for high-quality and diverse 3D scene generation. To expand a scene during generation, one needs only to append empty blocks to overlap with the current scene and extrapolate existing latent tri-planes to populate new blocks. The extrapolation is done by conditioning the generation process with the feature samples from the overlapping tri-planes during the denoising iterations. Latent tri-plane extrapolation produces semantically and geometrically meaningful transitions that harmoniously blend with the existing scene. A 2D layout conditioning mechanism is used to control the placement and arrangement of scene elements. Experimental results indicate that BlockFusion is capable of generating diverse, geometrically consistent and unbounded large 3D scenes with unprecedented high-quality shapes in both indoor and outdoor scenarios.

cs.CV↗

arXiv2024

Accelerating Decision Diagram-based Multi-node Quantum Simulation with Ring Communication and Automatic SWAP Insertion

An N-bit quantum state requires a vector of length $2^N$, leading to an exponential increase in the required memory with N in conventional statevector-based quantum simulators. A proposed solution to this issue is the decision diagram-based quantum simulator, which can significantly decrease the necessary memory and is expected to operate faster for specific quantum circuits. However, decision diagram-based quantum simulators are not easily parallelizable because data must be manipulated dynamically, and most implementations run on one thread. This paper introduces ring communication-based optimal parallelization and automatic swap insertion techniques for multi-node implementation of decision diagram-based quantum simulators. The ring communication approach is designed so that each node communicates with its neighboring nodes, which can facilitate faster and more parallel communication than broadcasting where one node needs to communicate with all nodes simultaneously. The automatic swap insertion method, an approach to minimize inter-node communication, has been employed in existing multi-node state vector-based simulators, but this paper proposes two methods specifically designed for decision diagram-based quantum simulators. These techniques were implemented and evaluated using the Shor algorithm and random circuits with up to 38 qubits using a maximum of 256 nodes. The experimental results have revealed that multi-node implementation can reduce run-time by up to 26 times. For example, Shor circuits that need 38 qubits can finish simulation in 147 seconds. Additionally, it was shown that ring communication has a higher speed-up effect than broadcast communication, and the importance of selecting the appropriate automatic swap insertion method was revealed.

quant-ph↗

arXiv2024

Hunting DeFi Vulnerabilities via Context-Sensitive Concolic Verification

Decentralized finance (DeFi) is revolutionizing the traditional centralized finance paradigm with its attractive features such as high availability, transparency, and tamper-proofing. However, attacks targeting DeFi services have severely damaged the DeFi market, as evidenced by our investigation of 80 real-world DeFi incidents from 2017 to 2022. Existing methods, based on symbolic execution, model checking, semantic analysis, and fuzzing, fall short in identifying the most DeFi vulnerability types. To address the deficiency, we propose Context-Sensitive Concolic Verification (CSCV), a method of automating the DeFi vulnerability finding based on user-defined properties formulated in temporal logic. CSCV builds and optimizes contexts to guide verification processes that dynamically construct context-carrying transition systems in tandem with concolic executions. Furthermore, we demonstrate the effectiveness of CSCV through experiments on real-world DeFi services and qualitative comparison. The experiment results show that our CSCV prototype successfully detects 76.25% of the vulnerabilities from the investigated incidents with an average time of 253.06 seconds.

cs.SE↗

arXiv2024

Dr.Hair: Reconstructing Scalp-Connected Hair Strands without Pre-training via Differentiable Rendering of Line Segments

In the film and gaming industries, achieving a realistic hair appearance typically involves the use of strands originating from the scalp. However, reconstructing these strands from observed surface images of hair presents significant challenges. The difficulty in acquiring Ground Truth (GT) data has led state-of-the-art learning-based methods to rely on pre-training with manually prepared synthetic CG data. This process is not only labor-intensive and costly but also introduces complications due to the domain gap when compared to real-world data. In this study, we propose an optimization-based approach that eliminates the need for pre-training. Our method represents hair strands as line segments growing from the scalp and optimizes them using a novel differentiable rendering algorithm. To robustly optimize a substantial number of slender explicit geometries, we introduce 3D orientation estimation utilizing global optimization, strand initialization based on Laplace's equation, and reparameterization that leverages geometric connectivity and spatial proximity. Unlike existing optimization-based methods, our method is capable of reconstructing internal hair flow in an absolute direction. Our method exhibits robust and accurate inverse rendering, surpassing the quality of existing methods and significantly improving processing speed.

cs.CV↗

arXiv2024

Self-organized clustering, prediction, and superposition of long-term cognitive decline from short-term individual cognitive test scores in Alzheimer's disease

Progressive cognitive decline spanning across decades is characteristic of Alzheimer's disease (AD). Various predictive models have been designed to realize its early onset and study the long-term trajectories of cognitive test scores across populations of interest. Research efforts have been geared towards superimposing patients' cognitive test scores with the long-term trajectory denoting gradual cognitive decline, while considering the heterogeneity of AD. Multiple trajectories representing cognitive assessment for the long-term have been developed based on various parameters, highlighting the importance of classifying several groups based on disease progression patterns. In this study, a novel method capable of self-organized prediction, classification, and the overlay of long-term cognitive trajectories based on short-term individual data was developed, based on statistical and differential equation modeling. We validated the predictive accuracy of the proposed method for the long-term trajectory of cognitive test score results on two cohorts: the Alzheimer's Disease Neuroimaging Initiative (ADNI) study and the Japanese ADNI study. We also presented two practical illustrations of the simultaneous evaluation of risk factor associated with both the onset and the longitudinal progression of AD, and an innovative randomized controlled trial design for AD that standardizes the heterogeneity of patients enrolled in a clinical trial. These resources would improve the power of statistical hypothesis testing and help evaluate the therapeutic effect. The application of predicting the trajectory of longitudinal disease progression goes beyond AD, and is especially relevant for progressive and neurodegenerative disorders.

q-bio.QM↗

arXiv2023

Parallelizing quantum simulation with decision diagrams

Recent technological advancements show promise in leveraging quantum mechanical phenomena for computation. This brings substantial speed-ups to problems that are once considered to be intractable in the classical world. However, the physical realization of quantum computers is still far away from us, and a majority of research work is done using quantum simulators running on classical computers. Classical computers face a critical obstacle in simulating quantum algorithms. Quantum states reside in a Hilbert space whose size grows exponentially to the number of subsystems, i.e., qubits. As a result, the straightforward statevector approach does not scale due to the exponential growth of the memory requirement. Decision diagrams have gained attention in recent years for representing quantum states and operations in quantum simulations. The main advantage of this approach is its ability to exploit redundancy. However, mainstream quantum simulators still rely on statevectors or tensor networks. We consider the absence of decision diagrams due to the lack of parallelization strategies. This work explores several strategies for parallelizing decision diagram operations, specifically for quantum simulations. We propose optimal parallelization strategies. Based on the experiment results, our parallelization strategy achieves a 2-3 times faster simulation of Grover's algorithm and random circuits than the state-of-the-art single-thread DD-based simulator DDSIM.

quant-ph↗