Selected work

Representative Papers

Epileptic Seizure Prediction Using Patient-Adaptive Transformer Networks

Mar 26, 2026

This study addresses the challenges of high inter-patient variability and complex temporal dynamics in electroencephalogram (EEG) signals for seizure prediction by proposing an adaptive Transformer framework tailored to individual patients. The approach employs a two-stage training strategy: first, a general-purpose EEG representation is learned via self-supervised pretraining; subsequently, patient-specific fine-tuning is performed by integrating noise-aware preprocessing, multi-channel signal tokenization, and autoregressive modeling. Evaluated on the TUH EEG dataset, the method achieves over 90% accuracy in predicting seizures within a 30-second horizon and attains an F1 score above 0.80, significantly outperforming existing approaches and demonstrating enhanced performance in personalized seizure forecasting.

0 citationsRead paper

PrompTrend: Continuous Community-Driven Vulnerability Discovery and Assessment for Large Language Models

Jul 25, 2025

Static benchmarking inadequately captures LLM security vulnerabilities exposed in real-world online community practices. To address this, we propose a community-driven dynamic monitoring paradigm, focusing on psychological attacks as the primary threat vector, revealing that capability advancement and safety improvement are misaligned. Methodologically, we design a cross-platform data collection system, a multidimensional scoring framework, and a scalable monitoring architecture—integrating horizontal comparative analysis with fine-grained vulnerability classification. Over five months, we conduct an empirical study across nine commercial LLMs. Our approach identifies 198 novel vulnerabilities with 78% classification accuracy; psychological attacks exhibit significantly higher detection rates than traditional exploit-based techniques yet demonstrate low cross-model transferability—highlighting their stealthiness and model specificity. This work pioneers systematic, continuous discovery and quantitative evaluation of community-emergent LLM vulnerabilities.

0 citationsRead paper

Bridging AI and Software Security: A Comparative Vulnerability Assessment of LLM Agent Deployment Paradigms

Jul 08, 2025

Current LLM agent security research treats AI-specific vulnerabilities and traditional software flaws in isolation, lacking a unified cross-domain evaluation framework. Method: This work presents the first systematic comparison of two dominant deployment paradigms—Function Calling (FC) and Model Context Protocol (MCP)—and introduces a unified threat taxonomy integrating AI reasoning vulnerabilities with classical software security concepts. We empirically evaluate the robustness of seven LLMs against prompt injection, JSON injection, DoS, and multi-step chained attacks across 3,250 adversarial scenarios. Contribution/Results: FC exhibits higher overall attack success (73.5% vs. 62.59%), but risks concentrate at the system layer; MCP exacerbates LLM centralization and exposure. Chained attacks achieve 91–96% success rates, and advanced reasoning models show heightened exploitability. Our findings reveal how architectural choices fundamentally reshape the threat landscape, establishing novel conceptual insights and practical benchmarks for secure LLM agent design.

0 citationsRead paper

Randomized Power Transmission With Optimized Level Selection Probabilities in Uncoordinated Uplink NOMA

Jul 01, 2025IEEE Wireless Communications Letters

This paper addresses the optimization of predefined power-level selection probabilities in uncoordinated uplink NOMA systems to minimize block error rate (BLER) or bit error rate (BER). Method: We propose a generic probabilistic optimization framework that is agnostic to specific multiuser detection algorithms and the number of power levels. It supports iterative solving under multiuser collision scenarios by jointly modeling randomized power allocation, channel statistics, and detection performance—thereby reformulating the problem as a tractable quadratic program and designing an efficient iterative algorithm. Contribution/Results: Experiments demonstrate that, under the assumption that the base station can concurrently demodulate signals from two or more users, the proposed method significantly reduces BLER/BER, enhances system access efficiency and robustness, and establishes a scalable probabilistic configuration paradigm for uncoordinated random-access NOMA.

0 citationsRead paper

Collision Resolution in RFID Systems Using Antenna Arrays and Mix Source Separation

May 01, 2025IEEE Communications Letters

To address tag identification failure caused by signal collisions in RFID systems, this paper proposes a novel pilot-free signal recovery method leveraging antenna arrays and blind source separation. The method introduces a hybrid objective function that jointly incorporates the Zero-Forcing Constant-Modulus (ZFCM) criterion and a newly designed ambiguity suppression criterion, effectively mitigating phase and amplitude ambiguities inherent in conventional ZFCM-based beamformer estimation; gradient descent optimization ensures efficient solution convergence. By fully exploiting the spatial resolution of antenna arrays, the approach achieves high-precision and robust separation of concurrent tag signals—without requiring channel state information, tag location knowledge, or pilot symbols. Experimental results demonstrate substantial improvements in identification accuracy under high-density tag scenarios, establishing a scalable physical-layer multiple-access solution for passive RFID systems.

0 citationsRead paper

Recent publications

Latest Papers

Epileptic Seizure Prediction Using Patient-Adaptive Transformer Networks

Mar 26, 2026

This study addresses the challenges of high inter-patient variability and complex temporal dynamics in electroencephalogram (EEG) signals for seizure prediction by proposing an adaptive Transformer framework tailored to individual patients. The approach employs a two-stage training strategy: first, a general-purpose EEG representation is learned via self-supervised pretraining; subsequently, patient-specific fine-tuning is performed by integrating noise-aware preprocessing, multi-channel signal tokenization, and autoregressive modeling. Evaluated on the TUH EEG dataset, the method achieves over 90% accuracy in predicting seizures within a 30-second horizon and attains an F1 score above 0.80, significantly outperforming existing approaches and demonstrating enhanced performance in personalized seizure forecasting.

0 citationsRead paper

PrompTrend: Continuous Community-Driven Vulnerability Discovery and Assessment for Large Language Models

Jul 25, 2025

Static benchmarking inadequately captures LLM security vulnerabilities exposed in real-world online community practices. To address this, we propose a community-driven dynamic monitoring paradigm, focusing on psychological attacks as the primary threat vector, revealing that capability advancement and safety improvement are misaligned. Methodologically, we design a cross-platform data collection system, a multidimensional scoring framework, and a scalable monitoring architecture—integrating horizontal comparative analysis with fine-grained vulnerability classification. Over five months, we conduct an empirical study across nine commercial LLMs. Our approach identifies 198 novel vulnerabilities with 78% classification accuracy; psychological attacks exhibit significantly higher detection rates than traditional exploit-based techniques yet demonstrate low cross-model transferability—highlighting their stealthiness and model specificity. This work pioneers systematic, continuous discovery and quantitative evaluation of community-emergent LLM vulnerabilities.

0 citationsRead paper

Bridging AI and Software Security: A Comparative Vulnerability Assessment of LLM Agent Deployment Paradigms

Jul 08, 2025

Current LLM agent security research treats AI-specific vulnerabilities and traditional software flaws in isolation, lacking a unified cross-domain evaluation framework. Method: This work presents the first systematic comparison of two dominant deployment paradigms—Function Calling (FC) and Model Context Protocol (MCP)—and introduces a unified threat taxonomy integrating AI reasoning vulnerabilities with classical software security concepts. We empirically evaluate the robustness of seven LLMs against prompt injection, JSON injection, DoS, and multi-step chained attacks across 3,250 adversarial scenarios. Contribution/Results: FC exhibits higher overall attack success (73.5% vs. 62.59%), but risks concentrate at the system layer; MCP exacerbates LLM centralization and exposure. Chained attacks achieve 91–96% success rates, and advanced reasoning models show heightened exploitability. Our findings reveal how architectural choices fundamentally reshape the threat landscape, establishing novel conceptual insights and practical benchmarks for secure LLM agent design.

0 citationsRead paper

Randomized Power Transmission With Optimized Level Selection Probabilities in Uncoordinated Uplink NOMA

Jul 01, 2025IEEE Wireless Communications Letters

This paper addresses the optimization of predefined power-level selection probabilities in uncoordinated uplink NOMA systems to minimize block error rate (BLER) or bit error rate (BER). Method: We propose a generic probabilistic optimization framework that is agnostic to specific multiuser detection algorithms and the number of power levels. It supports iterative solving under multiuser collision scenarios by jointly modeling randomized power allocation, channel statistics, and detection performance—thereby reformulating the problem as a tractable quadratic program and designing an efficient iterative algorithm. Contribution/Results: Experiments demonstrate that, under the assumption that the base station can concurrently demodulate signals from two or more users, the proposed method significantly reduces BLER/BER, enhances system access efficiency and robustness, and establishes a scalable probabilistic configuration paradigm for uncoordinated random-access NOMA.

0 citationsRead paper

Collision Resolution in RFID Systems Using Antenna Arrays and Mix Source Separation

May 01, 2025IEEE Communications Letters

To address tag identification failure caused by signal collisions in RFID systems, this paper proposes a novel pilot-free signal recovery method leveraging antenna arrays and blind source separation. The method introduces a hybrid objective function that jointly incorporates the Zero-Forcing Constant-Modulus (ZFCM) criterion and a newly designed ambiguity suppression criterion, effectively mitigating phase and amplitude ambiguities inherent in conventional ZFCM-based beamformer estimation; gradient descent optimization ensures efficient solution convergence. By fully exploiting the spatial resolution of antenna arrays, the approach achieves high-precision and robust separation of concurrent tag signals—without requiring channel state information, tag location knowledge, or pilot symbols. Experimental results demonstrate substantial improvements in identification accuracy under high-density tag scenarios, establishing a scalable physical-layer multiple-access solution for passive RFID systems.

0 citationsRead paper