Selected work

Representative Papers

Explore Simpler Eigenmarking: Quantum Entailment Model Checking

Apr 26, 2026

This work addresses the challenges of implementing Grover search for implication-based model checking, particularly the requirement of minority-state conditions and the difficulty of executing highly entangling operations on near-term quantum hardware. To overcome these limitations, the authors propose a simplified Eigenmarking scheme that employs only a single ancillary qubit and a universal two-qubit controlled-phase gate (CCZ), reducing the original multi-controlled phase rotation to a doubly controlled operation. This approach substantially diminishes reliance on high-entanglement states while preserving effective amplitude amplification and unsatisfiability detection capabilities. Consequently, it significantly alleviates hardware demands and enhances scalability. Simulation results demonstrate that the proposed method achieves markedly superior performance, with a minimum relative local winning rate of W = 3.17 and discriminability of D = 0.769, outperforming both conventional marking (W = 0.67, D = 0.19) and fine-grained marking (W = 0.28, D = 0.55).

0 citationsRead paper

Probabilistic Geometric Alignment via Bayesian Latent Transport for Domain-Adaptive Foundation Models

Mar 24, 2026

This work addresses the challenge of domain adaptation for foundation models, which often suffer from distributional shift, optimization instability, and poor uncertainty calibration in novel domains. The authors propose an uncertainty-aware alignment paradigm that formulates domain adaptation as a stochastic geometric alignment problem in representation space. By leveraging Bayesian latent variable transport along Wasserstein geodesics, the method redistributes probability mass while incorporating PAC-Bayesian regularization to control posterior complexity. This approach uniquely integrates stochastic optimal transport geometry with statistical generalization theory, simultaneously ensuring convergence stability and sample efficiency. It substantially reduces latent manifold discrepancy, accelerates transport energy decay, and enhances covariance calibration and cross-domain probabilistic reliability, outperforming both deterministic fine-tuning and adversarial domain adaptation methods.

0 citationsRead paper

A Real-Time Neuro-Symbolic Ethical Governor for Safe Decision Control in Autonomous Robotic Manipulation

Mar 15, 2026

This work addresses the lack of real-time, interpretable ethical oversight in autonomous robots operating in human-robot coexistence and safety-critical scenarios. It proposes the first neuro-symbolic framework for robotic ethical governance, integrating a fine-tuned DistilBERT model to parse ethical intent from natural language instructions, a probabilistic ethical risk field for modeling and uncertainty estimation, and a threshold-based override control mechanism to enable dynamic supervision and intervention in operational decisions. Experimental results demonstrate that the approach achieves stable convergence in simulated robotic arm tasks, effectively identifies ethical risks, and significantly enhances safe decision-making performance with minimal compromise to task efficiency, while also improving system transparency and interpretability.

0 citationsRead paper

Daily Fluctuations in Weather and Economic Growth at the Subnational Level: Evidence from Thailand

Jun 24, 2025

This study examines the historical impact of temperature variability on provincial per capita output growth in Thailand and projects future climate risks. Method: Employing a panel fixed-effects model, we integrate provincial economic and meteorological data from 1980–2019 and project impacts through 2090 under RCP4.5 and RCP8.5 scenarios. Contribution/Results: We identify a statistically significant inverted-U relationship between temperature and economic growth: a 1°C increase reduces annual per capita output growth by 1.25–3.80 percentage points, with effects concentrated in agriculture—industrial and service sectors show no significant response. This is the first provincial-scale quantification of climate change’s dynamic impact on long-run per capita income in Thailand: by 2090, 62%–86% of the population will experience relative declines in per capita income due to warming. Crucially, temperature primarily suppresses the *growth rate* of income rather than its static level—providing microeconometric evidence to inform climate-economic policy in Southeast Asia.

0 citationsRead paper

Deep Learning-Based Breast Cancer Detection in Mammography: A Multi-Center Validation Study in Thai Population

May 29, 2025

Addressing the urgent need for effective breast cancer screening in the Thai population, this study develops a clinically deployable, multi-center generalizable AI detection system. Method: We propose an enhanced EfficientNetV2 architecture incorporating a hybrid channel-spatial attention mechanism, integrated with lesion localization heatmap generation, multi-center data distribution adaptation during training, and a robustness evaluation framework for LLF/NLF localization. Contribution/Results: This work represents the first systematic validation of AI generalizability and radiologist-AI collaborative performance in real-world, multi-center Thai clinical settings, establishing a novel clinical concordance analysis paradigm. Experiments demonstrate an AUROC of 0.89–0.96, lesion localization LLF > 0.83, radiologist-AI concordance rates of 83.5% (classification) and 84.0% (localization), clinical acceptance > 89%, and usability score of 74.17 (source hospital), confirming high clinical applicability.

0 citationsRead paper

Recent publications

Latest Papers

Explore Simpler Eigenmarking: Quantum Entailment Model Checking

Apr 26, 2026

This work addresses the challenges of implementing Grover search for implication-based model checking, particularly the requirement of minority-state conditions and the difficulty of executing highly entangling operations on near-term quantum hardware. To overcome these limitations, the authors propose a simplified Eigenmarking scheme that employs only a single ancillary qubit and a universal two-qubit controlled-phase gate (CCZ), reducing the original multi-controlled phase rotation to a doubly controlled operation. This approach substantially diminishes reliance on high-entanglement states while preserving effective amplitude amplification and unsatisfiability detection capabilities. Consequently, it significantly alleviates hardware demands and enhances scalability. Simulation results demonstrate that the proposed method achieves markedly superior performance, with a minimum relative local winning rate of W = 3.17 and discriminability of D = 0.769, outperforming both conventional marking (W = 0.67, D = 0.19) and fine-grained marking (W = 0.28, D = 0.55).

0 citationsRead paper

Probabilistic Geometric Alignment via Bayesian Latent Transport for Domain-Adaptive Foundation Models

Mar 24, 2026

This work addresses the challenge of domain adaptation for foundation models, which often suffer from distributional shift, optimization instability, and poor uncertainty calibration in novel domains. The authors propose an uncertainty-aware alignment paradigm that formulates domain adaptation as a stochastic geometric alignment problem in representation space. By leveraging Bayesian latent variable transport along Wasserstein geodesics, the method redistributes probability mass while incorporating PAC-Bayesian regularization to control posterior complexity. This approach uniquely integrates stochastic optimal transport geometry with statistical generalization theory, simultaneously ensuring convergence stability and sample efficiency. It substantially reduces latent manifold discrepancy, accelerates transport energy decay, and enhances covariance calibration and cross-domain probabilistic reliability, outperforming both deterministic fine-tuning and adversarial domain adaptation methods.

0 citationsRead paper

A Real-Time Neuro-Symbolic Ethical Governor for Safe Decision Control in Autonomous Robotic Manipulation

Mar 15, 2026

This work addresses the lack of real-time, interpretable ethical oversight in autonomous robots operating in human-robot coexistence and safety-critical scenarios. It proposes the first neuro-symbolic framework for robotic ethical governance, integrating a fine-tuned DistilBERT model to parse ethical intent from natural language instructions, a probabilistic ethical risk field for modeling and uncertainty estimation, and a threshold-based override control mechanism to enable dynamic supervision and intervention in operational decisions. Experimental results demonstrate that the approach achieves stable convergence in simulated robotic arm tasks, effectively identifies ethical risks, and significantly enhances safe decision-making performance with minimal compromise to task efficiency, while also improving system transparency and interpretability.

0 citationsRead paper

Daily Fluctuations in Weather and Economic Growth at the Subnational Level: Evidence from Thailand

Jun 24, 2025

This study examines the historical impact of temperature variability on provincial per capita output growth in Thailand and projects future climate risks. Method: Employing a panel fixed-effects model, we integrate provincial economic and meteorological data from 1980–2019 and project impacts through 2090 under RCP4.5 and RCP8.5 scenarios. Contribution/Results: We identify a statistically significant inverted-U relationship between temperature and economic growth: a 1°C increase reduces annual per capita output growth by 1.25–3.80 percentage points, with effects concentrated in agriculture—industrial and service sectors show no significant response. This is the first provincial-scale quantification of climate change’s dynamic impact on long-run per capita income in Thailand: by 2090, 62%–86% of the population will experience relative declines in per capita income due to warming. Crucially, temperature primarily suppresses the *growth rate* of income rather than its static level—providing microeconometric evidence to inform climate-economic policy in Southeast Asia.

0 citationsRead paper

Deep Learning-Based Breast Cancer Detection in Mammography: A Multi-Center Validation Study in Thai Population

May 29, 2025

Addressing the urgent need for effective breast cancer screening in the Thai population, this study develops a clinically deployable, multi-center generalizable AI detection system. Method: We propose an enhanced EfficientNetV2 architecture incorporating a hybrid channel-spatial attention mechanism, integrated with lesion localization heatmap generation, multi-center data distribution adaptation during training, and a robustness evaluation framework for LLF/NLF localization. Contribution/Results: This work represents the first systematic validation of AI generalizability and radiologist-AI collaborative performance in real-world, multi-center Thai clinical settings, establishing a novel clinical concordance analysis paradigm. Experiments demonstrate an AUROC of 0.89–0.96, lesion localization LLF > 0.83, radiologist-AI concordance rates of 83.5% (classification) and 84.0% (localization), clinical acceptance > 89%, and usability score of 74.17 (source hospital), confirming high clinical applicability.

0 citationsRead paper