Selected work

Representative Papers

IceHorizon: A Dataset for Horizon Detection in Ice-Covered Maritime Environments and Comparative Evaluation of Detection Methods

Aug 07, 2026

This study addresses the challenges of horizon detection in icy maritime imagery, where low sea-sky contrast, ice clutter, and varying illumination conditions degrade performance. To this end, the authors introduce IceHorizon, the first dataset specifically designed for this scenario, comprising 30 shipborne and 8 UAV video sequences. They systematically evaluate six detection methods, including four classical computer vision algorithms and two hybrid approaches that integrate deep learning with traditional line-detection techniques. Experimental results demonstrate that the proposed hybrid methods achieve superior accuracy, robustness, and computational efficiency, with shipborne platforms significantly outperforming UAV-based ones. The dataset and source code have been publicly released to support future research in this domain.

0 citationsRead paper

Predictive Triggering for Outage-Resilient Threshold Decisions over Short-Packet Links

Aug 04, 2026

This work addresses the challenge of jointly achieving high reliability and energy efficiency in remote threshold-based decision-making over short-packet wireless links, where outage risk complicates the trade-off between timely decisions and power consumption. The authors propose a novel framework that integrates a predictive triggering mechanism with an age-of-information (AoI)-aware elastic update policy. By innovatively coupling prediction-based triggering with AoI control and modeling channel dynamics via a two-state Markov chain, the approach enables joint optimization of decision reliability and energy usage. Through a Bayesian posterior decision rule and co-design of transmit power and update probability, the method significantly advances decision timing at comparable energy costs while maintaining low false-alarm and miss-detection rates, outperforming existing baseline schemes.

0 citationsRead paper

Predicting Male Fertility Using Machine Learning: A Semen Parameters Based Analysis with the VISEM Dataset

Jul 09, 2026

Male infertility is often underdiagnosed due to the lack of objective assessment tools. This study systematically evaluates the performance of over forty machine learning models in classifying fertility status into three categories—fertile, subfertile, and infertile—using semen parameters (concentration, motility, and morphology) from the VISEM dataset comprising 85 subjects. Leveraging feature engineering, the LazyPredict automated modeling framework, five-fold cross-validation, and multiclass ROC-AUC analysis, the Nearest Centroid classifier emerged as the top-performing model, achieving an accuracy of 94.2%. Its performance significantly surpassed that of support vector machines and quadratic discriminant analysis, demonstrating strong potential as a clinical decision-support tool for male infertility diagnosis.

0 citationsRead paper

Riazi-8B: An Urdu Large Language Model for Mathematical Reasoning

Jun 24, 2026

This work addresses the scarcity of mathematical reasoning datasets and adapted models for low-resource languages such as Urdu, which severely limits the performance of large language models on multi-step math problems in such languages. The authors propose a two-stage adaptation strategy: first conducting continued pretraining on Urdu Wikipedia, followed by supervised fine-tuning using a Chain-of-Thought dataset constructed by translating GSM8K into Urdu. This approach uniquely integrates language adaptation with reasoning-oriented fine-tuning, effectively transferring mathematical reasoning capabilities to Urdu. Experimental results on the MGSM-Urdu benchmark demonstrate that the proposed method significantly outperforms existing Urdu instruction-tuned models, achieving consistent improvements in answer accuracy, reasoning quality, response completeness, and language generation fluency, thereby filling critical gaps in both data and modeling for mathematical reasoning in Urdu.

0 citationsRead paper

Lightweight Non-Line-of-Sight Channel Detection for ML-assisted Bluetooth Direction Finding

Jun 17, 2026

This study addresses the challenge of angle estimation bias in BLE direction finding under multipath environments, where reflections and scattering degrade accuracy, and the absence of lightweight line-of-sight (LOS)/non-line-of-sight (NLOS) detection methods suitable for narrowband signals. The authors construct a controlled BLE measurement setup to collect and annotate CTE IQ data, proposing a lightweight LOS/NLOS detection pipeline tailored for narrowband channels. Their approach leverages quantization normalization, principal component analysis (PCA), and adaptive kernel density estimation to extract robust features, and employs Nyström kernel approximation to construct a low-rank nonlinear mapping, which is then integrated with a support vector classifier for efficient discrimination. The proposed method achieves significantly reduced inference complexity and memory overhead while maintaining high accuracy, yielding a 7–14% relative improvement in accuracy over baseline feature representations and demonstrating superior suitability over models like MLPs in resource-constrained scenarios.

0 citationsRead paper

Recent publications

Latest Papers

IceHorizon: A Dataset for Horizon Detection in Ice-Covered Maritime Environments and Comparative Evaluation of Detection Methods

Aug 07, 2026

This study addresses the challenges of horizon detection in icy maritime imagery, where low sea-sky contrast, ice clutter, and varying illumination conditions degrade performance. To this end, the authors introduce IceHorizon, the first dataset specifically designed for this scenario, comprising 30 shipborne and 8 UAV video sequences. They systematically evaluate six detection methods, including four classical computer vision algorithms and two hybrid approaches that integrate deep learning with traditional line-detection techniques. Experimental results demonstrate that the proposed hybrid methods achieve superior accuracy, robustness, and computational efficiency, with shipborne platforms significantly outperforming UAV-based ones. The dataset and source code have been publicly released to support future research in this domain.

0 citationsRead paper

Predictive Triggering for Outage-Resilient Threshold Decisions over Short-Packet Links

Aug 04, 2026

This work addresses the challenge of jointly achieving high reliability and energy efficiency in remote threshold-based decision-making over short-packet wireless links, where outage risk complicates the trade-off between timely decisions and power consumption. The authors propose a novel framework that integrates a predictive triggering mechanism with an age-of-information (AoI)-aware elastic update policy. By innovatively coupling prediction-based triggering with AoI control and modeling channel dynamics via a two-state Markov chain, the approach enables joint optimization of decision reliability and energy usage. Through a Bayesian posterior decision rule and co-design of transmit power and update probability, the method significantly advances decision timing at comparable energy costs while maintaining low false-alarm and miss-detection rates, outperforming existing baseline schemes.

0 citationsRead paper

Predicting Male Fertility Using Machine Learning: A Semen Parameters Based Analysis with the VISEM Dataset

Jul 09, 2026

Male infertility is often underdiagnosed due to the lack of objective assessment tools. This study systematically evaluates the performance of over forty machine learning models in classifying fertility status into three categories—fertile, subfertile, and infertile—using semen parameters (concentration, motility, and morphology) from the VISEM dataset comprising 85 subjects. Leveraging feature engineering, the LazyPredict automated modeling framework, five-fold cross-validation, and multiclass ROC-AUC analysis, the Nearest Centroid classifier emerged as the top-performing model, achieving an accuracy of 94.2%. Its performance significantly surpassed that of support vector machines and quadratic discriminant analysis, demonstrating strong potential as a clinical decision-support tool for male infertility diagnosis.

0 citationsRead paper

Riazi-8B: An Urdu Large Language Model for Mathematical Reasoning

Jun 24, 2026

This work addresses the scarcity of mathematical reasoning datasets and adapted models for low-resource languages such as Urdu, which severely limits the performance of large language models on multi-step math problems in such languages. The authors propose a two-stage adaptation strategy: first conducting continued pretraining on Urdu Wikipedia, followed by supervised fine-tuning using a Chain-of-Thought dataset constructed by translating GSM8K into Urdu. This approach uniquely integrates language adaptation with reasoning-oriented fine-tuning, effectively transferring mathematical reasoning capabilities to Urdu. Experimental results on the MGSM-Urdu benchmark demonstrate that the proposed method significantly outperforms existing Urdu instruction-tuned models, achieving consistent improvements in answer accuracy, reasoning quality, response completeness, and language generation fluency, thereby filling critical gaps in both data and modeling for mathematical reasoning in Urdu.

0 citationsRead paper

Lightweight Non-Line-of-Sight Channel Detection for ML-assisted Bluetooth Direction Finding

Jun 17, 2026

This study addresses the challenge of angle estimation bias in BLE direction finding under multipath environments, where reflections and scattering degrade accuracy, and the absence of lightweight line-of-sight (LOS)/non-line-of-sight (NLOS) detection methods suitable for narrowband signals. The authors construct a controlled BLE measurement setup to collect and annotate CTE IQ data, proposing a lightweight LOS/NLOS detection pipeline tailored for narrowband channels. Their approach leverages quantization normalization, principal component analysis (PCA), and adaptive kernel density estimation to extract robust features, and employs Nyström kernel approximation to construct a low-rank nonlinear mapping, which is then integrated with a support vector classifier for efficient discrimination. The proposed method achieves significantly reduced inference complexity and memory overhead while maintaining high accuracy, yielding a 7–14% relative improvement in accuracy over baseline feature representations and demonstrating superior suitability over models like MLPs in resource-constrained scenarios.

0 citationsRead paper