Selected work

Representative Papers

Neural Fourier Surrogates for Data Reuploading Quantum Neural Networks

Sep 30, 2026

This study addresses the blurred boundary between quantum neural networks (QNNs) and classical models, which stems from the absence of fair comparison benchmarks. We propose the Neural Fourier Surrogate (NFS) architecture, which integrates neural quantum states with random Fourier feature techniques to efficiently learn coefficients under a unified Fourier basis. This approach constructs a classical surrogate network supporting finite Fourier series, serving as a natural classical baseline for evaluating data-reuploading QNNs. Experimental results demonstrate that NFS achieves performance comparable to both classical models and QNNs on tabular benchmarks. Consequently, this work establishes a rigorous and effective evaluation framework for quantifying potential quantum advantages in machine learning tasks.

0 citationsRead paper

ShamAN-Q: Shampoo Augmented NanoQuant for Sub-1-bit LLM Weights

Sep 29, 2026

This study addresses the significant accuracy degradation in sub-1-bit quantization of large language models caused by neglecting second-order curvature information. We introduce, for the first time, the dense curvature paradigm of the Shampoo optimizer into post-training quantization. Specifically, our method constructs a Kronecker-factored approximation of the empirical Fisher matrix via Kullback–Leibler divergence minimization and reformulates ADMM updates as Sylvester equations. Additionally, we propose layer-wise dynamic refreshing and cross-layer bit reallocation strategies. Evaluated on the Qwen3 series, our approach substantially reduces perplexity (e.g., from 27.56 to 22.96 for the 0.6B model), matching the performance of prior 1-bit methods at approximately 0.8 bits while preserving zero-shot accuracy.

0 citationsRead paper

Towards Scaling Quantum Fine-Tuning of Foundational Time Series Models for Classification

Sep 04, 2026

研究通过在时间序列模型Chronos上使用量子头进行微调，以解决电力网格事件分类问题，并引入wing模块来增加数据带宽，从而提高准确性。

0 citationsRead paper

Breakeven demonstration of quantum low-density parity-check codes

Jun 04, 2026

This work addresses the challenge of deploying high-rate quantum error-correcting codes, which are often hindered by hardware constraints such as long-range couplings. On a single trapped-ion quantum computer and without any hardware reconfiguration, the authors demonstrate, for the first time, flexible implementation of nine distinct error-correcting codes—spanning qLDPC, topological, and concatenated families—with markedly different connectivity requirements. Leveraging an optical–metastable–ground (OMG) architecture, the system enables addressable mid-circuit measurement and reset without requiring ion shuttling or dedicated coolant ions. Notably, a qLDPC code encoding four logical qubits into eighteen physical qubits achieves break-even performance, exhibiting a logical error rate nine times lower than comparable superconducting-platform experiments; moreover, certain logical qubits surpass the coherence time of their constituent physical qubits, substantially enhancing resource efficiency.

0 citationsRead paper

Quantum Parity Representations: Learnable Basis Discovery, Encoders, and Shadow Deployment

May 11, 2026

This work addresses the challenge of modeling high-order feature interactions under binarized or quantized inputs by proposing a classically efficient inference method that requires no quantum resources. Leveraging quantum-inspired ideas during training—through learnable Pauli word selection, projection-based encoding, and an sPQC-Parity architecture—the approach constructs parity representations that rely solely on classical computation at inference time. On native binary tasks with 5–10 bits, the method achieves accuracy improvements of 23.9%–41.7% over logistic regression and SVM, and significantly outperforms baselines such as PCA-bin on textual and discrete datasets. Notably, it even surpasses fully continuous models in certain scenarios, marking the first demonstration of a performance advantage for quantum-inspired parity representations within a classically efficient inference framework.

0 citationsRead paper

Recent publications

Latest Papers

Neural Fourier Surrogates for Data Reuploading Quantum Neural Networks

Sep 30, 2026

This study addresses the blurred boundary between quantum neural networks (QNNs) and classical models, which stems from the absence of fair comparison benchmarks. We propose the Neural Fourier Surrogate (NFS) architecture, which integrates neural quantum states with random Fourier feature techniques to efficiently learn coefficients under a unified Fourier basis. This approach constructs a classical surrogate network supporting finite Fourier series, serving as a natural classical baseline for evaluating data-reuploading QNNs. Experimental results demonstrate that NFS achieves performance comparable to both classical models and QNNs on tabular benchmarks. Consequently, this work establishes a rigorous and effective evaluation framework for quantifying potential quantum advantages in machine learning tasks.

0 citationsRead paper

ShamAN-Q: Shampoo Augmented NanoQuant for Sub-1-bit LLM Weights

Sep 29, 2026

This study addresses the significant accuracy degradation in sub-1-bit quantization of large language models caused by neglecting second-order curvature information. We introduce, for the first time, the dense curvature paradigm of the Shampoo optimizer into post-training quantization. Specifically, our method constructs a Kronecker-factored approximation of the empirical Fisher matrix via Kullback–Leibler divergence minimization and reformulates ADMM updates as Sylvester equations. Additionally, we propose layer-wise dynamic refreshing and cross-layer bit reallocation strategies. Evaluated on the Qwen3 series, our approach substantially reduces perplexity (e.g., from 27.56 to 22.96 for the 0.6B model), matching the performance of prior 1-bit methods at approximately 0.8 bits while preserving zero-shot accuracy.

0 citationsRead paper

Towards Scaling Quantum Fine-Tuning of Foundational Time Series Models for Classification

Sep 04, 2026

研究通过在时间序列模型Chronos上使用量子头进行微调，以解决电力网格事件分类问题，并引入wing模块来增加数据带宽，从而提高准确性。

0 citationsRead paper

Breakeven demonstration of quantum low-density parity-check codes

Jun 04, 2026

This work addresses the challenge of deploying high-rate quantum error-correcting codes, which are often hindered by hardware constraints such as long-range couplings. On a single trapped-ion quantum computer and without any hardware reconfiguration, the authors demonstrate, for the first time, flexible implementation of nine distinct error-correcting codes—spanning qLDPC, topological, and concatenated families—with markedly different connectivity requirements. Leveraging an optical–metastable–ground (OMG) architecture, the system enables addressable mid-circuit measurement and reset without requiring ion shuttling or dedicated coolant ions. Notably, a qLDPC code encoding four logical qubits into eighteen physical qubits achieves break-even performance, exhibiting a logical error rate nine times lower than comparable superconducting-platform experiments; moreover, certain logical qubits surpass the coherence time of their constituent physical qubits, substantially enhancing resource efficiency.

0 citationsRead paper

Quantum Parity Representations: Learnable Basis Discovery, Encoders, and Shadow Deployment

May 11, 2026

This work addresses the challenge of modeling high-order feature interactions under binarized or quantized inputs by proposing a classically efficient inference method that requires no quantum resources. Leveraging quantum-inspired ideas during training—through learnable Pauli word selection, projection-based encoding, and an sPQC-Parity architecture—the approach constructs parity representations that rely solely on classical computation at inference time. On native binary tasks with 5–10 bits, the method achieves accuracy improvements of 23.9%–41.7% over logistic regression and SVM, and significantly outperforms baselines such as PCA-bin on textual and discrete datasets. Notably, it even surpasses fully continuous models in certain scenarios, marking the first demonstration of a performance advantage for quantum-inspired parity representations within a classically efficient inference framework.

0 citationsRead paper