Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

IEEE International Conference on Quantum Computing and Engineering (QCE)

Generative Modeling of Measurement Data for Quantum Parameter Adaptation

Chen-Yu Liu, Samuel Yen-Chi Chen, Kuan‐Cheng Chen, Wei‐Jia Huang, Wei-Hao Huang, Yen-Jui Chang

Published Aug 30, 2025• 0 citations
Publisher

Abstract

Quantum Parameter Adaptation (QPA) has emerged as a promising approach for leveraging quantum neural networks (QNNs) to generate classical parameterefficient fine-tuning (PEFT) parameters, enabling quantum enhanced fine-tuning of large language models (LLMs). However, the practical implementation of QPA is hindered by the need for an extremely large number of quantum measurement shots, posing a significant challenge in realworld quantum computing environments. To address this issue, this work introduces Generative Interpolation (GI), a method inspired by frame generation in video data, where missing measurement probabilities are interpolated using a neural network-based generative model. By treating quantum measurement probabilities as analogous to video frames, GI estimates unmeasured basis state probabilities, significantly reducing the required quantum measurements. Empirical results demonstrate that incorporating GI into QPA reduces the quantum measurement shot requirement to just 2.5 % of the original count while achieving superior fine-tuning performance. This method not only enhances QPA efficiency but also establishes a broader connection between classical deep learning techniques and quantum measurement reconstruction. The proposed generative framework has the potential to extend to variational quantum algorithms, offering a pathway toward reducing quantum measurement overhead in hybrid quantum-classical computing paradigms.

DOI
10.1109/qce65121.2025.00187

Related papers

A Quantum Circuit-Based Compression Perspective for Parameter-Efficient Learning

Chen-Yu Liu, Chao-Han Huck Yang, Hsi‐Sheng Goan +1 more

Quantum-centric supercomputing presents a compelling framework for large-scale hybrid quantum-classical tasks. Although quantum machine learning (QML) offers theoretical benefits in various applications, challenges such as large-size data encoding in the input stage and the reliance on quantum resources in the inference stage limit its practicality for tasks like fine-tuning large language models (LLMs). Quantum parameter generation, a novel approach of QML, addresses these limitations by using quantum neural networks (QNNs) to generate classical model weights (parameters) exclusively during training, thereby decoupling inference from quantum hardware. In this work, we introduce Quantum Parameter Adaptation (QPA) in the framework of quantum parameter generation, which integrates QNNs with a classical multi-layer perceptron mapping model to generate parameters for fine-tuning methods. Using Gemma-2 and GPT-2 as case studies, QPA demonstrates significant parameter reduction for parameter-efficient fine-tuning methods, such as Low-Rank Adaptation (LoRA), while maintaining comparable or improved performance in text generation tasks. Specifically, QPA reduces the number of parameters to $52.06\%$ of the original LoRA for GPT-2 with a slight performance gain of $0.75\%$, and to $16.84\%$ for Gemma-2, with a marginal performance improvement of $0.07\%$. These results highlight QPA's ability to achieve efficient parameter reduction without sacrificing performance in the quantum parameter generation framework. This work showcases the potential of quantum-enhanced parameter reduction, offering a scalable quantum-classical solution for fine-tuning LLMs while preserving the feasibility of inference on classical hardware.

arXiv:2410.09846•Oct 13, 2024•15 citations

Hybrid Parameterized Quantum States for Variational Quantum Learning

Chen-Yu Liu

Variational quantum learning faces practical challenges in the noisy intermediate-scale quantum (NISQ) era. Parameterized quantum circuit (PQC) models suffer from statistical uncertainty due to finite-shot measurements and are highly sensitive to quantum noise, while purely classical approximations like neural quantum states (NQS) lack access to genuine quantum correlations and are limited in scalability. This work introduces Hybrid Parameterized Quantum States (HPQS), a general-purpose modeling framework that interpolates between quantum and classical parameterizations. HPQS combines PQC-based measurements with neural estimators via a blending mechanism and postprocessing functions, enabling enhanced, shot-efficient evaluation under hardware constraints. We demonstrate HPQS across three representative quantum learning tasks: (1) Expectation-based QML, where HPQS yields higher classification accuracy than PQC-only and NQS-only baselines under limited quantum measurements. (2) Quantum-Train, where HPQS generates the entire parameter set of classical networks using polylogarithmic trainable variables; and (3) Quantum Parameter Adaptation (QPA), where HPQS produces LoRA adapter parameters for fine-tuning large language models like GPT-2 and Gemma-2 with improved perplexity under low-shot conditions; Together, these results position HPQS as a scalable, noise-resilient approach for variational quantum learning, compatible with both current NISQ hardware and future fault-tolerant architectures.

arXiv:2505.16676•May 22, 2025•1 citation

Adaptive Quantum Computing Neural Network Architecture Optimization

Jincheng Zhang

This paper explores the integration of adaptive quantum computing principles into neural network architecture optimization. Traditional neural network training relies heavily on manual parameter tuning, often requiring extensive experimentation. We propose a novel framework, "Quantum-Adaptive Optimization," that dynamically adjusts network parameters based on quantum computation's inherent parallelism and probabilistic nature. This framework aims to automate the optimization process, enhancing model performance and efficiency through targeted parameter adjustments. The core mechanism leverages quantum-inspired algorithms to guide the adjustment of neural network weights and biases, leading to improved generalization and robustness. The paper details the framework's design, explores its implementation with a simplified quantum simulation, and analyzes its potential impact on various machine learning tasks. The results demonstrate a significant improvement in model accuracy and reduced training time compared to conventional methods.

•Sep 6, 2026•0 citations

How Can Quantum Deep Learning Improve Large Language Models?

Emily Jimin Roh, Hyojun Ahn, Samuel Yen-Chi Chen +2 more

The rapid progress of large language models (LLMs) has transformed natural language processing, yet the challenge of efficient adaptation remains unresolved. Full fine-tuning achieves strong performance but imposes prohibitive computational and memory costs. Parameter-efficient fine-tuning (PEFT) strategies, such as low-rank adaptation (LoRA), Prefix tuning, and sparse low-rank adaptation (SoRA), address this issue by reducing trainable parameters while maintaining competitive accuracy. However, these methods often encounter limitations in scalability, stability, and generalization across diverse tasks. Recent advances in quantum deep learning introduce novel opportunities through quantum-inspired encoding and parameterized quantum circuits (PQCs). In particular, the quantum-amplitude embedded adaptation (QAA) framework demonstrates expressive model updates with minimal overhead. This paper presents a systematic survey and comparative analysis of conventional PEFT methods and QAA. The analysis demonstrates trade-offs in convergence, efficiency, and representational capacity, while providing insight into the potential of quantum approaches for future LLM adaptation.

arXiv:2509.16244•IEEE International Conference on Acoustics Speech and Signal Processing•Apr 21, 2026•0 citations

Generative Learning for Quantum Measurement Design

Jun Dai, Olivier Nahman-Lévesque, Guillaume Rabusseau +2 more

Extracting quantum information from a quantum state is a fundamental task of quantum computation, often requiring the estimation of many non-commuting observables under a finite measurement budget. For both near-term and early fault-tolerant settings, the measurement protocol must balance statistical efficiency against implementation resources such as circuit depth, connectivity, and entangling-gate count. Many existing strategies focus on two extremes: hardware-friendly product measurements with high sampling cost, and fully commuting measurements with deep circuits. Here we recast resource-constrained measurement design as a generative learning problem. We introduce FlowMeas, which uses a generative flow network to directly sample finite ensembles of shallow Clifford measurement circuits subject to a prescribed shot budget and hardware constraints. At zero entangling depth, FlowMeas learns qubit-wise commuting measurement schedules and already matches or improves leading product-measurement methods on nearly all molecular benchmarks. Allowing one or two entangling gate layers yields further reductions in energy estimation error of up to $27\%$ relative to the strongest state-independent product-measurement baseline. The learned policy can also be reused across related Hamiltonians, substantially accelerating retraining along a molecular potential-energy surface. We further obtain results for molecular Hamiltonians with up to 20 qubits and apply the framework to a compactly encoded 54-qubit interacting fermionic model, extending the demonstrated scale beyond prior molecular benchmarks. These results establish generative learning as a flexible and unified framework for quantum measurement design under practical resource constraints.

quant-pharXiv:2608.11396•Aug 13, 2026•0 citations

Adaptive interpolating quantum transform: A quantum-native framework for efficient transform learning

Gekko Budiutama, Shunsuke Daimon, Hirofumi Nishi +3 more

Machine learning on quantum computers has attracted attention for its potential to deliver computational speedups in different tasks. However, deep variational quantum circuits require a large number of trainable parameters that grows with both qubit count and circuit depth, often rendering training infeasible. In this study we introduce the adaptive interpolating quantum transform (AIQT), a quantum-native framework for flexible and efficient learning. The AIQT defines a trainable unitary that interpolates between quantum transforms, such as the Hadamard and quantum Fourier transforms. This approach enables expressive quantum state manipulation while controlling parameter overhead. When built upon efficient quantum circuits such as the quantum Fourier transform, the AIQT achieves polynomial scaling of the number of gates with respect to the number of qubits. Our results show that the AIQT achieves high performance with minimal parameter count, offering a scalable and interpretable alternative to deep variational circuits.

arXiv:2508.14418•Physical Review A•Nov 6, 2025•1 citation
Quantum Parameter Adaptation (QPA) has emerged as a promising approach for leveraging quantum neural networks (QNNs) to generate classical parameterefficient fine-tuning (PEFT) parameters, enabling quantum enhanced fine-tuning of large language models (LLMs). However, the practical implementation of QPA is hindered by the need for an extremely large number of quantum measurement shots, posing a significant challenge in realworld quantum computing environments. To address this issue, this work introduces Generative Interpolation (GI), a method inspired by frame generation in video data, where missing measurement probabilities are interpolated using a neural network-based generative model. By treating quantum measurement probabilities as analogous to video frames, GI estimates unmeasured basis state probabilities, significantly reducing the required quantum measurements. Empirical results demonstrate that incorporating GI into QPA reduces the quantum measurement shot requirement to just 2.5 % of the original count while achieving superior fine-tuning performance. This method not only enhances QPA efficiency but also establishes a broader connection between classical deep learning techniques and quantum measurement reconstruction. The proposed generative framework has the potential to extend to variational quantum algorithms, offering a pathway toward reducing quantum measurement overhead in hybrid quantum-classical computing paradigms.
Quantum-centric supercomputing presents a compelling framework for large-scale hybrid quantum-classical tasks. Although quantum machine learning (QML) offers theoretical benefits in various applications, challenges such as large-size data encoding in the input stage and the reliance on quantum resources in the inference stage limit its practicality for tasks like fine-tuning large language models (LLMs). Quantum parameter generation, a novel approach of QML, addresses these limitations by using quantum neural networks (QNNs) to generate classical model weights (parameters) exclusively during training, thereby decoupling inference from quantum hardware. In this work, we introduce Quantum Parameter Adaptation (QPA) in the framework of quantum parameter generation, which integrates QNNs with a classical multi-layer perceptron mapping model to generate parameters for fine-tuning methods. Using Gemma-2 and GPT-2 as case studies, QPA demonstrates significant parameter reduction for parameter-efficient fine-tuning methods, such as Low-Rank Adaptation (LoRA), while maintaining comparable or improved performance in text generation tasks. Specifically, QPA reduces the number of parameters to 52.06%52.06\%52.06% of the original LoRA for GPT-2 with a slight performance gain of 0.75%0.75\%0.75%, and to 16.84%16.84\%16.84% for Gemma-2, with a marginal performance improvement of 0.07%0.07\%0.07%. These results highlight QPA's ability to achieve efficient parameter reduction without sacrificing performance in the quantum parameter generation framework. This work showcases the potential of quantum-enhanced parameter reduction, offering a scalable quantum-classical solution for fine-tuning LLMs while preserving the feasibility of inference on classical hardware.
Variational quantum learning faces practical challenges in the noisy intermediate-scale quantum (NISQ) era. Parameterized quantum circuit (PQC) models suffer from statistical uncertainty due to finite-shot measurements and are highly sensitive to quantum noise, while purely classical approximations like neural quantum states (NQS) lack access to genuine quantum correlations and are limited in scalability. This work introduces Hybrid Parameterized Quantum States (HPQS), a general-purpose modeling framework that interpolates between quantum and classical parameterizations. HPQS combines PQC-based measurements with neural estimators via a blending mechanism and postprocessing functions, enabling enhanced, shot-efficient evaluation under hardware constraints. We demonstrate HPQS across three representative quantum learning tasks: (1) Expectation-based QML, where HPQS yields higher classification accuracy than PQC-only and NQS-only baselines under limited quantum measurements. (2) Quantum-Train, where HPQS generates the entire parameter set of classical networks using polylogarithmic trainable variables; and (3) Quantum Parameter Adaptation (QPA), where HPQS produces LoRA adapter parameters for fine-tuning large language models like GPT-2 and Gemma-2 with improved perplexity under low-shot conditions; Together, these results position HPQS as a scalable, noise-resilient approach for variational quantum learning, compatible with both current NISQ hardware and future fault-tolerant architectures.
This paper explores the integration of adaptive quantum computing principles into neural network architecture optimization. Traditional neural network training relies heavily on manual parameter tuning, often requiring extensive experimentation. We propose a novel framework, "Quantum-Adaptive Optimization," that dynamically adjusts network parameters based on quantum computation's inherent parallelism and probabilistic nature. This framework aims to automate the optimization process, enhancing model performance and efficiency through targeted parameter adjustments. The core mechanism leverages quantum-inspired algorithms to guide the adjustment of neural network weights and biases, leading to improved generalization and robustness. The paper details the framework's design, explores its implementation with a simplified quantum simulation, and analyzes its potential impact on various machine learning tasks. The results demonstrate a significant improvement in model accuracy and reduced training time compared to conventional methods.
The rapid progress of large language models (LLMs) has transformed natural language processing, yet the challenge of efficient adaptation remains unresolved. Full fine-tuning achieves strong performance but imposes prohibitive computational and memory costs. Parameter-efficient fine-tuning (PEFT) strategies, such as low-rank adaptation (LoRA), Prefix tuning, and sparse low-rank adaptation (SoRA), address this issue by reducing trainable parameters while maintaining competitive accuracy. However, these methods often encounter limitations in scalability, stability, and generalization across diverse tasks. Recent advances in quantum deep learning introduce novel opportunities through quantum-inspired encoding and parameterized quantum circuits (PQCs). In particular, the quantum-amplitude embedded adaptation (QAA) framework demonstrates expressive model updates with minimal overhead. This paper presents a systematic survey and comparative analysis of conventional PEFT methods and QAA. The analysis demonstrates trade-offs in convergence, efficiency, and representational capacity, while providing insight into the potential of quantum approaches for future LLM adaptation.
Extracting quantum information from a quantum state is a fundamental task of quantum computation, often requiring the estimation of many non-commuting observables under a finite measurement budget. For both near-term and early fault-tolerant settings, the measurement protocol must balance statistical efficiency against implementation resources such as circuit depth, connectivity, and entangling-gate count. Many existing strategies focus on two extremes: hardware-friendly product measurements with high sampling cost, and fully commuting measurements with deep circuits. Here we recast resource-constrained measurement design as a generative learning problem. We introduce FlowMeas, which uses a generative flow network to directly sample finite ensembles of shallow Clifford measurement circuits subject to a prescribed shot budget and hardware constraints. At zero entangling depth, FlowMeas learns qubit-wise commuting measurement schedules and already matches or improves leading product-measurement methods on nearly all molecular benchmarks. Allowing one or two entangling gate layers yields further reductions in energy estimation error of up to 27%27\%27% relative to the strongest state-independent product-measurement baseline. The learned policy can also be reused across related Hamiltonians, substantially accelerating retraining along a molecular potential-energy surface. We further obtain results for molecular Hamiltonians with up to 20 qubits and apply the framework to a compactly encoded 54-qubit interacting fermionic model, extending the demonstrated scale beyond prior molecular benchmarks. These results establish generative learning as a flexible and unified framework for quantum measurement design under practical resource constraints.
Machine learning on quantum computers has attracted attention for its potential to deliver computational speedups in different tasks. However, deep variational quantum circuits require a large number of trainable parameters that grows with both qubit count and circuit depth, often rendering training infeasible. In this study we introduce the adaptive interpolating quantum transform (AIQT), a quantum-native framework for flexible and efficient learning. The AIQT defines a trainable unitary that interpolates between quantum transforms, such as the Hadamard and quantum Fourier transforms. This approach enables expressive quantum state manipulation while controlling parameter overhead. When built upon efficient quantum circuits such as the quantum Fourier transform, the AIQT achieves polynomial scaling of the number of gates with respect to the number of qubits. Our results show that the AIQT achieves high performance with minimal parameter count, offering a scalable and interpretable alternative to deep variational circuits.