Selected work

Representative Papers

Recent Advances in Medical Image Classification

Jun 04, 2025International Journal of Advanced Computer Science and Applications

Medical image classification faces two critical challenges: scarcity of labeled samples in clinical settings and insufficient interpretability for medical practitioners. This paper presents a systematic review of recent advances and proposes a novel few-shot learning framework that synergistically integrates Vision Transformers (ViTs) with vision-language models (VLMs)—the first such integration for medical image classification. To enhance clinical trustworthiness, the framework incorporates explainable AI (XAI) techniques to generate human-interpretable visualizations (e.g., attention heatmaps) and natural-language explanations aligned with clinical reasoning. Extensive experiments on multiple public medical imaging benchmarks demonstrate that our method significantly improves few-shot classification accuracy—achieving an average gain of +5.2%—while delivering clinically coherent, multimodal explanations. The approach advances the practical deployment of AI-assisted diagnosis by bridging the gap between high-performance deep learning and domain-specific interpretability requirements.

2 citationsRead paper

Implementation of a Metacognition Framework for Self-Awareness and Self-Regulation in Ensembles of LLMs

Aug 15, 2026

This study addresses the limitations of self-assessment and reliability in large language models (LLMs) by proposing a novel integrated metacognitive framework. We introduce a five-dimensional metacognitive state vector to quantify self-awareness, integrating dual-system theory with graph algorithms to enable adaptive switching between thinking modes and dynamic role allocation. As the first work to incorporate metacognitive mechanisms into LLM ensemble systems, this research validates precise routing control and process visualization capabilities. The proposed approach significantly enhances self-perception, conflict detection, and boundary recognition within LLMs. Ultimately, this framework establishes a new paradigm for improving the trustworthiness and reliability of AI systems by endowing them with structured metacognitive regulation.

0 citationsRead paper

Classical $\mathrm{SU}(2)$ Models Match or Exceed Shallow Variational Quantum Circuits on Vision Benchmarks

Aug 07, 2026

This study systematically evaluates the performance of classical and quantum models sharing an $\mathrm{SU}(2)$ geometric structure on visual recognition tasks, investigating whether shallow variational quantum circuits offer practical advantages. Building upon frozen features from a pretrained ResNet18 backbone, the authors compare real-valued, quaternion-based, and variational quantum classifiers—with and without entanglement—across MNIST, FashionMNIST, and CIFAR-10, optimizing all models using Fubini–Study natural gradients. The results demonstrate that quaternion networks match or closely approach real-valued baselines while significantly outperforming quantum counterparts; entanglement yields only marginal gains on grayscale images and degrades performance when applied to pretrained features. This work provides the first evidence that merely sharing an $\mathrm{SU}(2)$ structure is insufficient to confer a quantum advantage in such settings.

0 citationsRead paper

Hypercubes, Hyperplanes, and Constraint-Induced Complexity Collapse in Atomic Concept Learning

Aug 03, 2026

This study addresses the uneven distribution of instance-space complexity in higher-order atomic concept learning by introducing a locality-of-complexity perspective grounded in the geometric structures of hypercubes and hyperplanes. It reveals that logical complexity concentrates along the full diagonal, while complexity collapses on other hyperplanes due to constraint-induced simplifications. Through high-dimensional geometric modeling, analysis of logical equivalence classes, and construction of constrained hypothesis spaces—combined with canonical simple concepts, minimal orderings, and representative reduction mechanisms—the work systematically classifies the behavior of high-dimensional hyperplanes. The analysis fully resolves binary and ternary cases, characterizing properties of orthogonal families, partial diagonals, and full diagonals, and establishes an upper bound on the number of equivalence classes for non-full-diagonal hyperplanes that is independent of term depth.

0 citationsRead paper

A Longitudinal Study of Recently Observed Malicious Domains: Characteristics, Infrastructure, and Abuse Patterns

Jun 09, 2026

This study systematically characterizes the behavioral traits and infrastructure abuse patterns of 1.52 million malicious domains detected by VirusTotal between January and May 2026. Integrating multi-source data—including WHOIS records, passive DNS, and IP/ASN information—the analysis spans eight dimensions: domain lifecycle, registration concentration, bulk registration, brand impersonation, and others. The work reveals, at scale for the first time, that attackers rapidly activate domains within weeks of registration, heavily concentrate registrations among a few registrars and top-level domains, extensively leverage Cloudflare for domain fronting, and most frequently impersonate WhatsApp and Google. The project publicly releases a labeled dataset, significantly enhancing threat intelligence and defensive capabilities against malicious domains.

0 citationsRead paper

Recent publications

Latest Papers

Implementation of a Metacognition Framework for Self-Awareness and Self-Regulation in Ensembles of LLMs

Aug 15, 2026

This study addresses the limitations of self-assessment and reliability in large language models (LLMs) by proposing a novel integrated metacognitive framework. We introduce a five-dimensional metacognitive state vector to quantify self-awareness, integrating dual-system theory with graph algorithms to enable adaptive switching between thinking modes and dynamic role allocation. As the first work to incorporate metacognitive mechanisms into LLM ensemble systems, this research validates precise routing control and process visualization capabilities. The proposed approach significantly enhances self-perception, conflict detection, and boundary recognition within LLMs. Ultimately, this framework establishes a new paradigm for improving the trustworthiness and reliability of AI systems by endowing them with structured metacognitive regulation.

0 citationsRead paper

Classical $\mathrm{SU}(2)$ Models Match or Exceed Shallow Variational Quantum Circuits on Vision Benchmarks

Aug 07, 2026

This study systematically evaluates the performance of classical and quantum models sharing an $\mathrm{SU}(2)$ geometric structure on visual recognition tasks, investigating whether shallow variational quantum circuits offer practical advantages. Building upon frozen features from a pretrained ResNet18 backbone, the authors compare real-valued, quaternion-based, and variational quantum classifiers—with and without entanglement—across MNIST, FashionMNIST, and CIFAR-10, optimizing all models using Fubini–Study natural gradients. The results demonstrate that quaternion networks match or closely approach real-valued baselines while significantly outperforming quantum counterparts; entanglement yields only marginal gains on grayscale images and degrades performance when applied to pretrained features. This work provides the first evidence that merely sharing an $\mathrm{SU}(2)$ structure is insufficient to confer a quantum advantage in such settings.

0 citationsRead paper

Hypercubes, Hyperplanes, and Constraint-Induced Complexity Collapse in Atomic Concept Learning

Aug 03, 2026

This study addresses the uneven distribution of instance-space complexity in higher-order atomic concept learning by introducing a locality-of-complexity perspective grounded in the geometric structures of hypercubes and hyperplanes. It reveals that logical complexity concentrates along the full diagonal, while complexity collapses on other hyperplanes due to constraint-induced simplifications. Through high-dimensional geometric modeling, analysis of logical equivalence classes, and construction of constrained hypothesis spaces—combined with canonical simple concepts, minimal orderings, and representative reduction mechanisms—the work systematically classifies the behavior of high-dimensional hyperplanes. The analysis fully resolves binary and ternary cases, characterizing properties of orthogonal families, partial diagonals, and full diagonals, and establishes an upper bound on the number of equivalence classes for non-full-diagonal hyperplanes that is independent of term depth.

0 citationsRead paper

A Longitudinal Study of Recently Observed Malicious Domains: Characteristics, Infrastructure, and Abuse Patterns

Jun 09, 2026

This study systematically characterizes the behavioral traits and infrastructure abuse patterns of 1.52 million malicious domains detected by VirusTotal between January and May 2026. Integrating multi-source data—including WHOIS records, passive DNS, and IP/ASN information—the analysis spans eight dimensions: domain lifecycle, registration concentration, bulk registration, brand impersonation, and others. The work reveals, at scale for the first time, that attackers rapidly activate domains within weeks of registration, heavily concentrate registrations among a few registrars and top-level domains, extensively leverage Cloudflare for domain fronting, and most frequently impersonate WhatsApp and Google. The project publicly releases a labeled dataset, significantly enhancing threat intelligence and defensive capabilities against malicious domains.

0 citationsRead paper

Towards the Development of Detection of Learned Helplessness in Mathematics: Design and Data Collection Challenges from a Developing Country Perspective

Apr 27, 2026

This study addresses the challenge of effectively collecting student interaction data in resource-constrained classrooms of developing countries, where such limitations hinder the identification of learned helplessness behaviors in mathematics learning. To overcome this, the authors developed a web-based linear equation tutoring system featuring adaptive problem sequencing, multimodal gamification mechanisms, and cross-platform compatibility. By logging interactions such as problem skipping, hint usage, and difficulty progression, the system enables the construction of models for detecting learned helplessness. Designed specifically for low-resource educational settings, it successfully mitigates real-world constraints—including outdated devices, unstable internet connectivity, and frequent instructional interruptions—and collected high-quality interaction data from 118 students out of an initial cohort of 410. The deployment reveals critical data collection bottlenecks while offering a replicable framework for educational data gathering and behavioral modeling in similar environments.

0 citationsRead paper