Selected work

Representative Papers

Exact solutions to the Weighted Region Problem

Feb 19, 2024arXiv.org

This paper investigates the exact computability of shortest paths in weighted rectangular domains. In the rational algebraic computation model, we establish—for the first time—that the globally shortest path in a single rectangular domain with piecewise nonnegative weights (where path cost equals Euclidean length multiplied by weight) is algorithmically undecidable. Method: For source points located either on the boundary or in the interior, we explicitly construct and derive algebraic equations for bisectors in the shortest path map (SPM); their coefficients are rational functions of the input parameters. Leveraging algebraic computation theory, implicit curve analysis, and structural characterization of SPMs, we develop a complete analytic framework for exact shortest paths. Results: Our work rigorously delineates the boundary of exact solvability for this problem and provides the first bisector computation framework implementable within the rational algebraic model.

2 citationsRead paper

Emergent 3D Instance Segmentation from Self-Supervised Point Transformers

Aug 16, 2026

This study addresses the reliance on handcrafted geometric priors in unsupervised 3D instance segmentation for outdoor LiDAR by proposing TokenGraph3D. Leveraging the observation that attention keys in self-supervised point Transformers encode instance signals under Rotary Position Embedding (RoPE), this method constructs a key similarity graph and applies connected component analysis to achieve training-free graph grouping without density clustering priors. Experiments demonstrate that TokenGraph3D significantly outperforms feature-based baselines in prior-free settings, successfully enabling the emergence of 3D instance structures across multiple datasets. Consequently, this approach effectively overcomes the limitations of traditional methods dependent on manually designed rules, offering a robust solution for unsupervised outdoor scene understanding.

0 citationsRead paper

Traceable LLM-Generated Hazard Scenarios for Operational Safety Analysis of Aviation Systems Using ASRS Reports

Aug 05, 2026

This study addresses the challenge of analyzing complex, multidimensional safety risks in aviation systems, which traditional functional hazard analysis methods struggle to capture comprehensively. To overcome this limitation, the authors propose a novel, traceable, and structured approach for generating hypothetical hazard scenarios using large language models (LLMs). The method automatically constructs coherent hazard narratives from NASA Aviation Safety Reporting System (ASRS) reports and evaluates their plausibility through historical co-occurrence evidence. Innovatively integrating evolutionary abduction with a hybrid generation mechanism—combining zero-shot and few-shot prompting alongside optional fine-tuning—the framework employs evolutionary algorithms to optimize both structural validity and narrative consistency. Experimental results demonstrate that this hybrid strategy significantly enhances the realism, logical correctness, and diversity of generated scenarios, outperforming approaches based on single-generation paradigms.

0 citationsRead paper

Beyond the Gegenbauer Paradigm: q-Orthogonal Kernels for Machine Learning

Aug 04, 2026

This work proposes a novel q-orthogonal kernel function by introducing discrete q-Hermite I polynomials into support vector machine (SVM) kernel design—a first in the literature. The proposed kernel leverages the deformation parameter \( q \) to inherently mitigate numerical underflow and overflow issues without requiring explicit scaling, thereby addressing the longstanding trade-off among interpretability, numerical stability, and computational efficiency that plagues existing orthogonal polynomial kernels. The constructed kernel satisfies Mercer’s condition, ensuring theoretical soundness while maintaining numerical robustness and computational simplicity. Furthermore, it provides a foundation for quantum-inspired algorithms. Empirical evaluation across 20 benchmark datasets demonstrates that the new kernel consistently matches or outperforms both classical and state-of-the-art orthogonal polynomial kernels, confirming its practical efficacy and potential.

0 citationsRead paper

A Distributed Acoustic Sensing Dataset for Vessel Detection and Localization in Submarine Cable Protection

Jul 30, 2026

This study addresses the frequent damage to subsea communication and power cables caused by vessel activities, highlighting the urgent need for effective monitoring solutions. To this end, the authors construct and release the Marlinks-NS dataset, which comprises ten days of continuous acoustic signals recorded via distributed acoustic sensing (DAS) over a 2,554-meter segment of a 28-kilometer buried fiber-optic cable in the North Sea. Integrated with Automatic Identification System (AIS) vessel data, the dataset supports two benchmark tasks: vessel detection and vessel-to-cable distance estimation. It includes 74,771 annotated samples across 250 sensing channels, representing the first large-scale, real-world marine DAS dataset with ground-truth labels. Provided in HDF5 format with spectral energy features, anonymized distance labels, and example code, Marlinks-NS enables reproducible research and lays a foundational data resource for intelligent cable protection systems.

0 citationsRead paper

Recent publications

Latest Papers

Emergent 3D Instance Segmentation from Self-Supervised Point Transformers

Aug 16, 2026

This study addresses the reliance on handcrafted geometric priors in unsupervised 3D instance segmentation for outdoor LiDAR by proposing TokenGraph3D. Leveraging the observation that attention keys in self-supervised point Transformers encode instance signals under Rotary Position Embedding (RoPE), this method constructs a key similarity graph and applies connected component analysis to achieve training-free graph grouping without density clustering priors. Experiments demonstrate that TokenGraph3D significantly outperforms feature-based baselines in prior-free settings, successfully enabling the emergence of 3D instance structures across multiple datasets. Consequently, this approach effectively overcomes the limitations of traditional methods dependent on manually designed rules, offering a robust solution for unsupervised outdoor scene understanding.

0 citationsRead paper

Traceable LLM-Generated Hazard Scenarios for Operational Safety Analysis of Aviation Systems Using ASRS Reports

Aug 05, 2026

This study addresses the challenge of analyzing complex, multidimensional safety risks in aviation systems, which traditional functional hazard analysis methods struggle to capture comprehensively. To overcome this limitation, the authors propose a novel, traceable, and structured approach for generating hypothetical hazard scenarios using large language models (LLMs). The method automatically constructs coherent hazard narratives from NASA Aviation Safety Reporting System (ASRS) reports and evaluates their plausibility through historical co-occurrence evidence. Innovatively integrating evolutionary abduction with a hybrid generation mechanism—combining zero-shot and few-shot prompting alongside optional fine-tuning—the framework employs evolutionary algorithms to optimize both structural validity and narrative consistency. Experimental results demonstrate that this hybrid strategy significantly enhances the realism, logical correctness, and diversity of generated scenarios, outperforming approaches based on single-generation paradigms.

0 citationsRead paper

Beyond the Gegenbauer Paradigm: q-Orthogonal Kernels for Machine Learning

Aug 04, 2026

This work proposes a novel q-orthogonal kernel function by introducing discrete q-Hermite I polynomials into support vector machine (SVM) kernel design—a first in the literature. The proposed kernel leverages the deformation parameter \( q \) to inherently mitigate numerical underflow and overflow issues without requiring explicit scaling, thereby addressing the longstanding trade-off among interpretability, numerical stability, and computational efficiency that plagues existing orthogonal polynomial kernels. The constructed kernel satisfies Mercer’s condition, ensuring theoretical soundness while maintaining numerical robustness and computational simplicity. Furthermore, it provides a foundation for quantum-inspired algorithms. Empirical evaluation across 20 benchmark datasets demonstrates that the new kernel consistently matches or outperforms both classical and state-of-the-art orthogonal polynomial kernels, confirming its practical efficacy and potential.

0 citationsRead paper

A Distributed Acoustic Sensing Dataset for Vessel Detection and Localization in Submarine Cable Protection

Jul 30, 2026

This study addresses the frequent damage to subsea communication and power cables caused by vessel activities, highlighting the urgent need for effective monitoring solutions. To this end, the authors construct and release the Marlinks-NS dataset, which comprises ten days of continuous acoustic signals recorded via distributed acoustic sensing (DAS) over a 2,554-meter segment of a 28-kilometer buried fiber-optic cable in the North Sea. Integrated with Automatic Identification System (AIS) vessel data, the dataset supports two benchmark tasks: vessel detection and vessel-to-cable distance estimation. It includes 74,771 annotated samples across 250 sensing channels, representing the first large-scale, real-world marine DAS dataset with ground-truth labels. Provided in HDF5 format with spectral energy features, anonymized distance labels, and example code, Marlinks-NS enables reproducible research and lays a foundational data resource for intelligent cable protection systems.

0 citationsRead paper

Developing and Validating the Spanish Version of the Large Language Models Dependency Scale (LLM-D12-SP)

Jul 24, 2026

This study addresses the lack of culturally adapted and reliable instruments for assessing psychological dependence on large language models (LLMs) among Spanish-speaking populations by developing and validating a Spanish version of the LLM Dependence Scale (LLM-D12). Confirmatory factor analysis, reliability testing (Cronbach’s α), and analyses of discriminant and external validity confirm that the scale exhibits a stable two-factor structure—comprising instrumental and relational dependence—with strong overall internal consistency (α = 0.89). The scale also demonstrates significant associations with internet addiction and perceived LLM trustworthiness. This work represents the first cross-cultural validation of an LLM dependence measure in a non-English-speaking population, offering a valid and generalizable tool for investigating the psychological aspects of LLM use in multilingual contexts.

0 citationsRead paper