Selected work

Representative Papers

Benchmarking Quantum and Classical Sequential Models for Urban Telecommunication Forecasting

Aug 06, 2025

This study addresses univariate SMS-in traffic forecasting on the Milan Telecom dataset, a critical urban communication flow prediction task. Method: We systematically evaluate five sequence modeling paradigms—classical LSTM and four quantum-inspired models (QLSTM, QASA, QRWKV, QFWP)—across varying input lengths (4–64). Contribution/Results: Contrary to the “quantum advantage” assumption, quantum-enhanced models do not universally outperform classical LSTM; their superiority is highly contingent on task characteristics, architectural design, and sequence length. Notably, models exhibit markedly divergent sensitivity to input length, exposing fundamental trade-offs among model capacity, parameter efficiency, and long-horizon temporal modeling capability. These findings empirically challenge the oversimplified notion of inherent quantum superiority and provide methodological insights and evidence-based guidance for rational architecture selection and structural optimization of quantum-inspired models in large-scale urban traffic forecasting.

0 citationsRead paper

Quantum Reinforcement Learning Trading Agent for Sector Rotation in the Taiwan Stock Market

Jun 25, 2025

This study addresses the agent-objective mismatch in quantum reinforcement learning (QRL) for financial decision-making—characterized by high training rewards but poor out-of-sample performance—by proposing a hybrid quantum-classical RL framework for sector rotation in the Taiwan stock market. Methodologically, it adopts Proximal Policy Optimization (PPO) as the backbone, integrating LSTM and Transformer architectures with three quantum-enhanced modules: Quantum Neural Networks (QNNs), Quantum-Enhanced RWKV (QRWKV), and Quantum-Adaptive Self-Attention (QASA), alongside automated feature engineering, under NISQ-device constraints for end-to-end training. Contributions include: (1) establishing the first reproducible quantum-classical financial RL benchmark; (2) empirically demonstrating that short-horizon reward functions induce overfitting, resulting in significantly lower cumulative returns and Sharpe ratios versus classical baselines; and (3) identifying limited quantum circuit expressivity and optimization instability as key bottlenecks underlying the quantum-classical performance gap.

0 citationsRead paper

Vision-QRWKV: Exploring Quantum-Enhanced RWKV Models for Image Classification

Jun 07, 2025

To address the limited nonlinear expressivity of high-dimensional visual features in image classification, this work introduces, for the first time, a quantum-enhanced mechanism into the RWKV architecture by embedding a variational quantum circuit (VQC) into its channel-mixing module, yielding a lightweight quantum-classical hybrid image classifier. We propose an end-to-end differentiable hybrid training framework jointly implemented using PyTorch and PennyLane. Systematic evaluation across 14 benchmark datasets—including the MedMNIST suite—demonstrates that our model achieves average accuracy gains of 2.3–4.1% on medical imaging tasks (e.g., ChestMNIST, RetinaMNIST, BloodMNIST), significantly improving discrimination of subtle visual differences and noisy samples, while outperforming Transformer-based models of comparable scale in inference efficiency. This work establishes the first systematic application of quantum-enhanced RWKV to vision tasks and provides a novel paradigm for efficient visual modeling under resource constraints.

0 citationsRead paper

Quantum Generative Models for Image Generation: Insights from MNIST and MedMNIST

Mar 30, 2025

Quantum generative models remain in their infancy, lacking systematic methodologies and efficient mechanisms for leveraging quantum noise. Method: This paper proposes a hybrid quantum generative framework for image synthesis, wherein the generator is implemented via a variational quantum circuit. It innovatively integrates intrinsic hardware quantum noise with controllable temporal noise scheduling to enable synergistic modeling; further, it introduces an adaptive noise injection strategy embedded within a hybrid quantum-classical training architecture. Contribution/Results: Evaluated on MNIST and MedMNIST, the model substantially outperforms existing quantum generative baselines. Empirical results demonstrate that quantum noise can be deliberately engineered and regulated—not only enhancing generation fidelity but also establishing noise as a learnable resource, thereby pioneering a new paradigm in quantum generative modeling.

0 citationsRead paper

Recent publications

Latest Papers

Benchmarking Quantum and Classical Sequential Models for Urban Telecommunication Forecasting

Aug 06, 2025

This study addresses univariate SMS-in traffic forecasting on the Milan Telecom dataset, a critical urban communication flow prediction task. Method: We systematically evaluate five sequence modeling paradigms—classical LSTM and four quantum-inspired models (QLSTM, QASA, QRWKV, QFWP)—across varying input lengths (4–64). Contribution/Results: Contrary to the “quantum advantage” assumption, quantum-enhanced models do not universally outperform classical LSTM; their superiority is highly contingent on task characteristics, architectural design, and sequence length. Notably, models exhibit markedly divergent sensitivity to input length, exposing fundamental trade-offs among model capacity, parameter efficiency, and long-horizon temporal modeling capability. These findings empirically challenge the oversimplified notion of inherent quantum superiority and provide methodological insights and evidence-based guidance for rational architecture selection and structural optimization of quantum-inspired models in large-scale urban traffic forecasting.

0 citationsRead paper

Quantum Reinforcement Learning Trading Agent for Sector Rotation in the Taiwan Stock Market

Jun 25, 2025

This study addresses the agent-objective mismatch in quantum reinforcement learning (QRL) for financial decision-making—characterized by high training rewards but poor out-of-sample performance—by proposing a hybrid quantum-classical RL framework for sector rotation in the Taiwan stock market. Methodologically, it adopts Proximal Policy Optimization (PPO) as the backbone, integrating LSTM and Transformer architectures with three quantum-enhanced modules: Quantum Neural Networks (QNNs), Quantum-Enhanced RWKV (QRWKV), and Quantum-Adaptive Self-Attention (QASA), alongside automated feature engineering, under NISQ-device constraints for end-to-end training. Contributions include: (1) establishing the first reproducible quantum-classical financial RL benchmark; (2) empirically demonstrating that short-horizon reward functions induce overfitting, resulting in significantly lower cumulative returns and Sharpe ratios versus classical baselines; and (3) identifying limited quantum circuit expressivity and optimization instability as key bottlenecks underlying the quantum-classical performance gap.

0 citationsRead paper

Vision-QRWKV: Exploring Quantum-Enhanced RWKV Models for Image Classification

Jun 07, 2025

To address the limited nonlinear expressivity of high-dimensional visual features in image classification, this work introduces, for the first time, a quantum-enhanced mechanism into the RWKV architecture by embedding a variational quantum circuit (VQC) into its channel-mixing module, yielding a lightweight quantum-classical hybrid image classifier. We propose an end-to-end differentiable hybrid training framework jointly implemented using PyTorch and PennyLane. Systematic evaluation across 14 benchmark datasets—including the MedMNIST suite—demonstrates that our model achieves average accuracy gains of 2.3–4.1% on medical imaging tasks (e.g., ChestMNIST, RetinaMNIST, BloodMNIST), significantly improving discrimination of subtle visual differences and noisy samples, while outperforming Transformer-based models of comparable scale in inference efficiency. This work establishes the first systematic application of quantum-enhanced RWKV to vision tasks and provides a novel paradigm for efficient visual modeling under resource constraints.

0 citationsRead paper

Quantum Generative Models for Image Generation: Insights from MNIST and MedMNIST

Mar 30, 2025

Quantum generative models remain in their infancy, lacking systematic methodologies and efficient mechanisms for leveraging quantum noise. Method: This paper proposes a hybrid quantum generative framework for image synthesis, wherein the generator is implemented via a variational quantum circuit. It innovatively integrates intrinsic hardware quantum noise with controllable temporal noise scheduling to enable synergistic modeling; further, it introduces an adaptive noise injection strategy embedded within a hybrid quantum-classical training architecture. Contribution/Results: Evaluated on MNIST and MedMNIST, the model substantially outperforms existing quantum generative baselines. Empirical results demonstrate that quantum noise can be deliberately engineered and regulated—not only enhancing generation fidelity but also establishing noise as a learnable resource, thereby pioneering a new paradigm in quantum generative modeling.

0 citationsRead paper