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

arXiv:2505.13205

Quantum Knowledge Distillation for Large Language Models

Lingxiao Li, Wang, Yihao, Jiacheng Fan, Li, Jing, Su‐Juan Qin, Qiaoyan Wen, Fei Gao

Published May 19, 2025• 24 citations
arXivHTMLPDF

Abstract

As foundational tools in natural language processing, Large Language Models (LLMs) have immense parameter scales, which makes deployment and inference increasingly prohibitive, especially in resource-constrained devices. Therefore, knowledge distillation for LLMs, i.e., compressing the LLM to a smaller model, is meaningful. With strong parameter representation capacity, quantum computing is regarded as a promising solution. Here, we propose a Quantum knowledge Distillation model for LLMs (QD-LLM) that leverages variational quantum circuits to learn from LLMs. In classical simulation, QD-LLM outperforms several mainstream distillation methods on multiple text classification tasks in terms of both accuracy and efficiency using only 11 qubits. The results reveal an interesting phenomenon that the simulation of quantum student models may be regarded as a new class of quantum-inspired classical algorithms. Remarkably, we deploy the obtained circuits on the Baihua superconducting quantum processor via the Quafu platform to assess practical feasibility. The model maintains stable inference performance despite hardware constraints such as decoherence and finite sampling. In summary, QD-LLM marks a foundational step in connecting quantum computing with LLMs, demonstrating the feasibility of quantum-native approaches that aim to compress and deploy models of increasingly larger scales. The code of this article has been open-sourced at https://github.com/Lilingxiao-bupt/QD-LLM.

DOI
10.48550/arxiv.2505.13205

Related papers

Quantum Knowledge Distillation: State-of-the-art, Applications and Open Problems

Aldana Al-Sulaiti, Kuan‐Cheng Chen, Muhammad Bilal Akram Dastagir +2 more

Knowledge distillation transfers predictions, representations, or other supervisory information between learning models and is increasingly investigated in quantum machine learning. Its applications extend beyond compression to cross-architecture transfer, circuit approximation, distributed learning, and classical deployment of models trained using quantum-derived supervision. This survey reviews these approaches using two complementary classification axes: teacher-student configuration and the role of quantum resources in the distillation process. Quantum-inspired methods are discussed separately from methods that simulate quantum computations classically or execute them on quantum hardware. We explain how measurement probabilities, expectation values, classical readouts, state overlaps, and circuit-level objectives support different transfer mechanisms. We then critically assess the presented results against baseline design, execution conditions, uncertainty, and resource costs. The evidence shows that distillation can improve selected quantum students, but benefits are task- and architecture-dependent, and several studies find negligible gains or even performance degradation. Evaluations on physical quantum hardware remain limited, with some testing individual quantum components rather than complete learning pipelines. Heterogeneous protocols and incomplete resource reporting prevent general conclusions about practical quantum advantage. We identify priorities for controlled benchmarking, noise-aware learning, reproducible evaluation, and efficient transfer between quantum and classical models.

arXiv:2611.08461•Nov 5, 2026•0 citations

Distilling the knowledge with quantum neural networks

Yuxuan Yan, Sitian Qian, Qi Zhao +1 more

Quantum Neural Networks (QNNs) are a promising class of quantum machine learning models with potential quantum advantages when implemented on scalable, error-corrected quantum computers. However, as system sizes increase, deploying QNNs becomes challenging. Similar to their classical counterparts, a key obstacle to their practical applications is that large-scale QNNs may not be easily deployed on smaller systems that have limited resources. Here, we tackle this challenge by compressing QNNs via knowledge distillation. We demonstrate how well-trained QNNs on large systems can be distilled into smaller architectures with similar configurations. We numerically show that knowledge distillation helps reduce the training cost of QNNs in terms of the number of qubits and circuit depth. Additionally, we find that a self-knowledge-distillation approach can accelerate training convergence. We believe our results offer new strategies for the efficient compression and practical deployment of QNNs.

quant-pharXiv:2603.21586•Mar 24, 2026•2 citations

Quantum-enhanced hybrid-model compression using knowledge distillation

Luigi , Massimo Esposito, Francesco Gargiulo

Abstract Quantum computing has emerged as a promising paradigm for addressing computational tasks intractable for classical systems, leveraging quantum mechanical principles such as superposition and entanglement to efficiently explore high-dimensional solution spaces. In recent years, hybrid quantum-classical approaches have gained increasing attention as a means to exploit the representational power of quantum systems while preserving the practicality of classical machine learning frameworks. This paper presents a study on quantum-enhanced model compression via feature-based knowledge distillation, in which a large classical neural network (the teacher) transfers knowledge to a substantially smaller hybrid student model. The proposed hybrid quantum-classical student architecture incorporates a parameterized quantum circuit (PQC) implemented in Qiskit, employing a linear $$\varvec{R_y}$$ angle-encoding feature map followed by an EfficientSU2 variational ansatz, with features bounded through a $$\varvec{\tanh (\cdot )\times \frac{\pi }{2}}$$ scaling to ensure stable gradient propagation. Gradient computation is performed via Reverse-Mode (adjoint) differentiation through the StatevectorEstimator primitive, which provides exact, shot-noise-free gradients at the cost of a single forward–backward pass per optimization step. The hybrid student is compared against a parameter-matched purely classical student under strict 1:1 parameter parity across a six-task binary benchmark spanning MNIST and Fashion-MNIST, as well as a four-class multi-class benchmark, totaling eleven classification experiments. A hardware validation experiment is additionally conducted on the IQM Garnet 20-qubit quantum processor using the Parameter-Shift Rule. The results indicate that the hybrid student outperforms its classical counterpart in five of six binary tasks (83%) and in the two hardest multi-class configurations, with advantages most pronounced in high-confusion regimes where compact classical classifiers approach the limits of their expressive capacity. On real quantum hardware, accuracy decreases from 99.04% in noiseless simulation to 80.50%, reflecting the impact of gate noise, readout errors, and transpilation overhead. While the hybrid model demonstrates improved parameter efficiency and expressive capacity, it incurs higher computational overhead and exhibits greater epoch-to-epoch training variability relative to classical baselines. We discuss practical limitations related to noise, scalability, and current hardware constraints, and identify directions for future work including hardware-efficient gradient estimation and advanced error mitigation protocols. Overall, our findings suggest that variational quantum circuits can serve as compact yet expressive components within neural compression schemes, with their representational advantage most effectively realized in binary and high-ambiguity classification scenarios.

Quantum Machine Intelligence•Sep 4, 2026•0 citations

Code-level quantum circuit generation based on large language models

HE Zhimin, Guohong Li, Haozhen Situ +3 more

Large language models (LLMs), based on deep learning techniques, are trained on vast amounts of textual data and possess the ability to understand and generate natural language.They have found widespread application in tasks such as machine translation, text generation, and question answering.This work proposes a novel approach for code-level quantum circuit generation using LLM. By fine-tuning existing LLM, we enable it to generate quantum circuit code tailored to user specifications, offering an efficient and high-performance solution for quantum circuit design. Simulation results demonstrate that this method achieves high accuracy in generating quantum circuit codes with only a small set of training data. This work provides a new perspective on quantum circuit design. It highlights the potential of LLMs in quantum programming and algorithm automation, laying the foundation for future research and practical applications in quantum artificial intelligence.

Zhongguo kexue. Wulixue Lixue Tianwenxue•Feb 18, 2025•2 citations

GroverGPT: A Large Language Model with 8 Billion Parameters for Quantum Searching

Haoran Wang, Pingzhi Li, Min Chen +3 more

Abstract Quantum computing is an exciting non-Von Neu-mann paradigm, offering provable speedups over classical computing for specific problems. However, the practical limits of classical simulatability for quantum circuits remain unclear, especially with current noisy quantum devices. In this work, we explore the potential of leveraging Large Language Models (LLMs) to simulate the output of a quantum Turing machine using Grover’s quantum circuits, known to provide quadratic speedups over classical counterparts. To this end, we developed GroverGPT, a specialized model based on LLaMA’s 8-billion-parameter architecture, trained on over 15 trillion tokens. Unlike brute-force state-vector simulations, which demand substantial computational resources, GroverGPT employs pattern recognition to approximate quantum search algorithms without explicitly representing quantum states. Analyzing 97K quantum search instances, GroverGPT consistently outperformed OpenAI’s GPT-4o (45% accuracy), achieving nearly 100% accuracy on 6-and 10-qubit datasets when trained on 4-qubit or larger datasets. It also demonstrated strong generalization, surpassing 95% accuracy for systems with over 20 qubits when trained on 3-to 6-qubit data. Analysis indicates GroverGPT captures quantum features of Grover’s search rather than classical patterns, supported by novel prompting strategies to enhance performance. Although accuracy declines with increasing system size, these findings offer insights into the practical boundaries of classical simulatability. This work suggests task-specific LLMs can surpass general-purpose models like GPT-4o in quantum algorithm learning and serve as powerful tools for advancing quantum research.

arXiv:2501.00135•Mar 6, 2026•15 citations

LLM-QFL: Distilling Large Language Model for Quantum Federated Learning

Dev Gurung, Shiva Raj Pokhrel

As Quantum Federated Learning (QFL) scales toward distributed quantum networks, managing heterogeneous resources and communication bottlenecks becomes a critical challenge. This research proposes LLM-QFL, an adaptive network service management framework that leverages Large Language Models (LLMs) to optimize the operational efficiency of QFL systems. We introduce a federated distillation method in which locally fine-tuned LLMs serve as autonomous network agents. These agents adaptively manage service parameters by: i) dynamically adjusting local computation intensity (optimizer steps) based on loss gradients, ii) performing variance-aware client selection to minimize network-wide heterogeneity, and iii) implementing intelligent early stopping criteria to conserve bandwidth. By serving as an orchestration layer, LLM-QFL provides a synergy between LLMs and quantum networking. Our contributions include: i) Adaptive Performance and Efficiency: Reducing idle computation and significantly cutting communication overhead; ii) Theoretical Rigor: Convergence guarantees of O(1/T) for the adaptive management protocol; and iii) Scalable Deployment: Implementing PEFT (LoRA/QLoRA) for resource-constrained quantum service nodes.

arXiv:2505.18656•IEEE Transactions on Network and Service Management•Jan 1, 2026•2 citations
As foundational tools in natural language processing, Large Language Models (LLMs) have immense parameter scales, which makes deployment and inference increasingly prohibitive, especially in resource-constrained devices. Therefore, knowledge distillation for LLMs, i.e., compressing the LLM to a smaller model, is meaningful. With strong parameter representation capacity, quantum computing is regarded as a promising solution. Here, we propose a Quantum knowledge Distillation model for LLMs (QD-LLM) that leverages variational quantum circuits to learn from LLMs. In classical simulation, QD-LLM outperforms several mainstream distillation methods on multiple text classification tasks in terms of both accuracy and efficiency using only 11 qubits. The results reveal an interesting phenomenon that the simulation of quantum student models may be regarded as a new class of quantum-inspired classical algorithms. Remarkably, we deploy the obtained circuits on the Baihua superconducting quantum processor via the Quafu platform to assess practical feasibility. The model maintains stable inference performance despite hardware constraints such as decoherence and finite sampling. In summary, QD-LLM marks a foundational step in connecting quantum computing with LLMs, demonstrating the feasibility of quantum-native approaches that aim to compress and deploy models of increasingly larger scales. The code of this article has been open-sourced at https://github.com/Lilingxiao-bupt/QD-LLM.
Knowledge distillation transfers predictions, representations, or other supervisory information between learning models and is increasingly investigated in quantum machine learning. Its applications extend beyond compression to cross-architecture transfer, circuit approximation, distributed learning, and classical deployment of models trained using quantum-derived supervision. This survey reviews these approaches using two complementary classification axes: teacher-student configuration and the role of quantum resources in the distillation process. Quantum-inspired methods are discussed separately from methods that simulate quantum computations classically or execute them on quantum hardware. We explain how measurement probabilities, expectation values, classical readouts, state overlaps, and circuit-level objectives support different transfer mechanisms. We then critically assess the presented results against baseline design, execution conditions, uncertainty, and resource costs. The evidence shows that distillation can improve selected quantum students, but benefits are task- and architecture-dependent, and several studies find negligible gains or even performance degradation. Evaluations on physical quantum hardware remain limited, with some testing individual quantum components rather than complete learning pipelines. Heterogeneous protocols and incomplete resource reporting prevent general conclusions about practical quantum advantage. We identify priorities for controlled benchmarking, noise-aware learning, reproducible evaluation, and efficient transfer between quantum and classical models.
Quantum Neural Networks (QNNs) are a promising class of quantum machine learning models with potential quantum advantages when implemented on scalable, error-corrected quantum computers. However, as system sizes increase, deploying QNNs becomes challenging. Similar to their classical counterparts, a key obstacle to their practical applications is that large-scale QNNs may not be easily deployed on smaller systems that have limited resources. Here, we tackle this challenge by compressing QNNs via knowledge distillation. We demonstrate how well-trained QNNs on large systems can be distilled into smaller architectures with similar configurations. We numerically show that knowledge distillation helps reduce the training cost of QNNs in terms of the number of qubits and circuit depth. Additionally, we find that a self-knowledge-distillation approach can accelerate training convergence. We believe our results offer new strategies for the efficient compression and practical deployment of QNNs.
Abstract Quantum computing has emerged as a promising paradigm for addressing computational tasks intractable for classical systems, leveraging quantum mechanical principles such as superposition and entanglement to efficiently explore high-dimensional solution spaces. In recent years, hybrid quantum-classical approaches have gained increasing attention as a means to exploit the representational power of quantum systems while preserving the practicality of classical machine learning frameworks. This paper presents a study on quantum-enhanced model compression via feature-based knowledge distillation, in which a large classical neural network (the teacher) transfers knowledge to a substantially smaller hybrid student model. The proposed hybrid quantum-classical student architecture incorporates a parameterized quantum circuit (PQC) implemented in Qiskit, employing a linear $$\varvec{R_y}$$ angle-encoding feature map followed by an EfficientSU2 variational ansatz, with features bounded through a $$\varvec{\tanh (\cdot )\times \frac{\pi }{2}}$$ scaling to ensure stable gradient propagation. Gradient computation is performed via Reverse-Mode (adjoint) differentiation through the StatevectorEstimator primitive, which provides exact, shot-noise-free gradients at the cost of a single forward–backward pass per optimization step. The hybrid student is compared against a parameter-matched purely classical student under strict 1:1 parameter parity across a six-task binary benchmark spanning MNIST and Fashion-MNIST, as well as a four-class multi-class benchmark, totaling eleven classification experiments. A hardware validation experiment is additionally conducted on the IQM Garnet 20-qubit quantum processor using the Parameter-Shift Rule. The results indicate that the hybrid student outperforms its classical counterpart in five of six binary tasks (83%) and in the two hardest multi-class configurations, with advantages most pronounced in high-confusion regimes where compact classical classifiers approach the limits of their expressive capacity. On real quantum hardware, accuracy decreases from 99.04% in noiseless simulation to 80.50%, reflecting the impact of gate noise, readout errors, and transpilation overhead. While the hybrid model demonstrates improved parameter efficiency and expressive capacity, it incurs higher computational overhead and exhibits greater epoch-to-epoch training variability relative to classical baselines. We discuss practical limitations related to noise, scalability, and current hardware constraints, and identify directions for future work including hardware-efficient gradient estimation and advanced error mitigation protocols. Overall, our findings suggest that variational quantum circuits can serve as compact yet expressive components within neural compression schemes, with their representational advantage most effectively realized in binary and high-ambiguity classification scenarios.
Large language models (LLMs), based on deep learning techniques, are trained on vast amounts of textual data and possess the ability to understand and generate natural language.They have found widespread application in tasks such as machine translation, text generation, and question answering.This work proposes a novel approach for code-level quantum circuit generation using LLM. By fine-tuning existing LLM, we enable it to generate quantum circuit code tailored to user specifications, offering an efficient and high-performance solution for quantum circuit design. Simulation results demonstrate that this method achieves high accuracy in generating quantum circuit codes with only a small set of training data. This work provides a new perspective on quantum circuit design. It highlights the potential of LLMs in quantum programming and algorithm automation, laying the foundation for future research and practical applications in quantum artificial intelligence.
Abstract Quantum computing is an exciting non-Von Neu-mann paradigm, offering provable speedups over classical computing for specific problems. However, the practical limits of classical simulatability for quantum circuits remain unclear, especially with current noisy quantum devices. In this work, we explore the potential of leveraging Large Language Models (LLMs) to simulate the output of a quantum Turing machine using Grover’s quantum circuits, known to provide quadratic speedups over classical counterparts. To this end, we developed GroverGPT, a specialized model based on LLaMA’s 8-billion-parameter architecture, trained on over 15 trillion tokens. Unlike brute-force state-vector simulations, which demand substantial computational resources, GroverGPT employs pattern recognition to approximate quantum search algorithms without explicitly representing quantum states. Analyzing 97K quantum search instances, GroverGPT consistently outperformed OpenAI’s GPT-4o (45% accuracy), achieving nearly 100% accuracy on 6-and 10-qubit datasets when trained on 4-qubit or larger datasets. It also demonstrated strong generalization, surpassing 95% accuracy for systems with over 20 qubits when trained on 3-to 6-qubit data. Analysis indicates GroverGPT captures quantum features of Grover’s search rather than classical patterns, supported by novel prompting strategies to enhance performance. Although accuracy declines with increasing system size, these findings offer insights into the practical boundaries of classical simulatability. This work suggests task-specific LLMs can surpass general-purpose models like GPT-4o in quantum algorithm learning and serve as powerful tools for advancing quantum research.
As Quantum Federated Learning (QFL) scales toward distributed quantum networks, managing heterogeneous resources and communication bottlenecks becomes a critical challenge. This research proposes LLM-QFL, an adaptive network service management framework that leverages Large Language Models (LLMs) to optimize the operational efficiency of QFL systems. We introduce a federated distillation method in which locally fine-tuned LLMs serve as autonomous network agents. These agents adaptively manage service parameters by: i) dynamically adjusting local computation intensity (optimizer steps) based on loss gradients, ii) performing variance-aware client selection to minimize network-wide heterogeneity, and iii) implementing intelligent early stopping criteria to conserve bandwidth. By serving as an orchestration layer, LLM-QFL provides a synergy between LLMs and quantum networking. Our contributions include: i) Adaptive Performance and Efficiency: Reducing idle computation and significantly cutting communication overhead; ii) Theoretical Rigor: Convergence guarantees of O(1/T) for the adaptive management protocol; and iii) Scalable Deployment: Implementing PEFT (LoRA/QLoRA) for resource-constrained quantum service nodes.