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

Asia Conference on Machine Learning and Computing (ACMLC)

Accelerating Molecular Generation with Structure-Aware Quantum Diffusion Models

Zisen Shan, Lianghong Chen, Pingzhao Hu

Published Jul 1, 2026• 0 citations
Publisher

Abstract

The high computational cost of classical diffusion models can limit their use in large-scale 3D molecular generation for drug and material discovery. We introduce Structure-aware Quantum Diffusion (SQ-Diff), the first full quantum diffusion model for this task, designed to leverage potential quantum advantages in the Noisy Intermediate-Scale Quantum era. Structural priors (e.g., inter-atomic distances) are encoded into the initial quantum state via a novel state preparation procedure that yields a unified normalization scheme dependent only on the number of atoms. The denoising process is driven by a Quantum U-Net, a fully quantum architecture that combines learnable variational quantum circuits with parameter-free operators. Training is guided by these structural priors enforced through a graph-based objective function to maintain structural consistency. Experimentally, SQ-Diff generates valid and diverse 3D molecules and shows improved performance over existing quantum-based methods. While a gap in generation quality compared to leading classical models remains, our model matches the inference speed of the fastest classical approaches with only a few quantum parameters, advancing the state of the art for pure quantum generative models.

DOI
10.1109/acmlc70381.2026.11700465

Related papers

Quantum Diffusion Model for Quark and Gluon Jet Generation

Mariia Baidachna, Rey Guadarrama, Gopal Ramesh Dahale +6 more

Diffusion models have demonstrated remarkable success in image generation, but they are computationally intensive and time-consuming to train. In this paper, we introduce a novel diffusion model that benefits from quantum computing techniques in order to mitigate computational challenges and enhance generative performance within high energy physics data. The fully quantum diffusion model replaces Gaussian noise with random unitary matrices in the forward process and incorporates a variational quantum circuit within the U-Net in the denoising architecture. We run evaluations on the structurally complex quark and gluon jets dataset from the Large Hadron Collider. The results demonstrate that the fully quantum and hybrid models are competitive with a similar classical model for jet generation, highlighting the potential of using quantum techniques for machine learning problems.

arXiv:2412.21082•Proceedings of the AAAI Symposium Series•Nov 23, 2025•5 citations

Quantum Hybrid Diffusion Models for Image Synthesis

Francesca De Falco, Andrea Ceschini, Alessandro Sebastianelli +2 more

Abstract In this paper, we propose a new methodology to design quantum hybrid diffusion models, derived from classical U-Nets with ResNet and Attention layers. Specifically, we propose two possible different hybridization schemes combining quantum computing’s superior generalization with classical networks’ modularity. In the first one, we acted at the vertex: ResNet convolutional layers are gradually replaced with variational circuits to create Quantum ResNet blocks. In the second proposed architecture, we extend the hybridization to the intermediate level of the encoder, due to its higher sensitivity in the feature extraction process. In order to conduct an in-depth analysis of the potential advantages stemming from the integration of quantum layers, images generated by quantum hybrid diffusion models are compared to those generated by classical models, and evaluated in terms of several quantitative metrics. The results demonstrate an advantage in using hybrid quantum diffusion models, as they generally synthesize better-quality images and converges faster. Moreover, they show the additional advantage of having a lower number of parameters to train compared to the classical one, with a reduction that depends on the extent to which the vertex is hybridized.

arXiv:2402.16147•Künstliche Intell.•Aug 9, 2024•29 citations

Quantum Generative Diffusion Model for Real-World Time Series

Jack Waller, Filippo Caruso, Dimitrios Makris +2 more

Generative models have achieved remarkable success in data synthesis, though recent advances driven by increasing model scale have introduced challenges in computational cost and efficiency. Quantum machine learning offers a promising alternative, representing complex data distributions using compact, highly expressive models. Here, we propose QDiffusion-TS, the first quantum generative diffusion model for time series synthesis, and validate it on the IQM quantum processor. The framework extends a classical diffusion architecture by replacing feed-forward components within the denoising transformer with quantum neural networks, yielding a hybrid quantum transformer that reduces the number of trainable parameters in each replaced component by nearly three orders of magnitude. Evaluated on financial time series from Apple and Amazon, the model generates synthetic data that more accurately reproduces the real distributions, reducing Wasserstein distance by approximately 44% relative to its classical counterpart across both datasets. In a downstream forecasting task, augmentation with the generated data improves predictive performance by up to 71% in RMSE over a baseline trained solely on real data. These results show that quantum enhanced architectures can consistently match and frequently surpass classical performance with substantially fewer parameters, establishing a practical framework towards more efficient and scalable data-driven generative modelling.

cs.LGCross-listedarXiv:2606.27561•Jun 29, 2026•3 citations

Quantum Diffusion Models in Earth Observation

F. Mauro, F. De Falco, L. Papa +6 more

Quantum diffusion models are rapidly advancing generative remote-sensing analytics by harnessing quantum computation to deliver higher-quality, more realistic satellite images, while converging faster and with substantially fewer trainable parameters than comparable classical diffusers. In this context, we first realise a fully-quantum latent diffusion model (QLDM) whose denoiser is a variational-quantum circuit acting on a 10-dimensionaI latent code. On EuroSAT, focusing on three land-cover classes Forest, Herbaceous Vegetation and SeaLake, QLDM lowers the Fréchet Inception Distance (FID) and Kernel Inception Distance (KID) by 21.5 % and 29.9 %, respectively, relative to a parameter-matched classical latent diffuser. Yet the severe spatial compression intrinsic to latent processing limits fine-grained detail. To restore image fidelity, we develop a hybrid quantum—classical architecture—the Quanvolutional Conditioned U-Net (QCU-Net)—which inserts entangling quantum layers both at the U-Net bottleneck and as a quanvolutional filter early in the encoder. Trained on the full ten-class EuroSAT RGB set, QCU-Net reaches an FID of 2.57 and a KID of 0.0008, representing 64 % and 76 % improvements over the best classical diffusion baseline (FID = 7.22; KID = 0.0034), and boosts class-conditioning accuracy to 81.7 % versus 62.2 % for the classical model. Thèse gains confirm that embedding quantum circuits within the feature-extraction pipeline yields richer spatial— spectral representations than latent-space quantum processing alone. Our results outline a clear progression in quantum generative modelling for Earth Observation: by first demonstrating with the QLDM that fully quantum denoising is not only feasible but also highly parameter- efficient, then showing how the hybrid QCU-Net architecture restores perceptual fidelity by embedding quantum feature extractors within a classical U-Net backbone, and finally pointing toward a fully pixel-space quantum diffusion model as future work—one that would leverage deep quantum circuits to perform both the forward and reverse diffusion directly in Hilbert space, remove classical auto-encoding steps, and promise even greater image quality and efficiency on fault-tolerant hardware.

•Jan 1, 2025•0 citations

Generative Quantum Machine Learning via Denoising Diffusion Probabilistic Models

Bingzhi Zhang, Peng Xu, Xiaohui Chen +1 more

Deep generative models are key-enabling technology to computer vision, text generation, and large language models. Denoising diffusion probabilistic models (DDPMs) have recently gained much attention due to their ability to generate diverse and high-quality samples in many computer vision tasks, as well as to incorporate flexible model architectures and a relatively simple training scheme. Quantum generative models, empowered by entanglement and superposition, have brought new insight to learning classical and quantum data. Inspired by the classical counterpart, we propose the quantum denoising diffusion probabilistic model (QuDDPM) to enable efficiently trainable generative learning of quantum data. QuDDPM adopts sufficient layers of circuits to guarantee expressivity, while it introduces multiple intermediate training tasks as interpolation between the target distribution and noise to avoid barren plateau and guarantee efficient training. We provide bounds on the learning error and demonstrate QuDDPM's capability in learning correlated quantum noise model, quantum many-body phases, and topological structure of quantum data. The results provide a paradigm for versatile and efficient quantum generative learning.

arXiv:2310.05866•Physical Review Letters•Mar 5, 2024•75 citations

Hybrid AI-Quantum Models for Real-Time Molecular Structure Prediction

Prof. (Dr) MSR Prasad

Hybrid artificial intelligence–quantum (AI–Q) models arise at the intersection of two rapidly advancing computational paradigms, promising unprecedented capabilities for real-time molecular structure prediction. In classical computational chemistry, the accurate determination of ground-state energies and optimized geometries relies on ab initio methods such as density functional theory (DFT) and coupled-cluster approaches, which, while highly precise, exhibit steep polynomial or exponential scaling with system size. Conversely, deep neural networks (DNNs) have demonstrated remarkable success in learning potential energy surfaces (PES) from large datasets, achieving rapid inference but sometimes lacking the ultimate chemical accuracy required for predictive applications in drug discovery or materials design. On the other hand, variational quantum eigensolvers (VQEs) leverage the intrinsic parallelism of quantum hardware to approximate electronic structure problems, yet they remain constrained by qubit coherence times, gate fidelities, and circuit depth limitations inherent to noisy intermediate-scale quantum (NISQ) devices.

World Journal of Future Technologies in Computer Science and Engineering•Jun 9, 2026•0 citations
The high computational cost of classical diffusion models can limit their use in large-scale 3D molecular generation for drug and material discovery. We introduce Structure-aware Quantum Diffusion (SQ-Diff), the first full quantum diffusion model for this task, designed to leverage potential quantum advantages in the Noisy Intermediate-Scale Quantum era. Structural priors (e.g., inter-atomic distances) are encoded into the initial quantum state via a novel state preparation procedure that yields a unified normalization scheme dependent only on the number of atoms. The denoising process is driven by a Quantum U-Net, a fully quantum architecture that combines learnable variational quantum circuits with parameter-free operators. Training is guided by these structural priors enforced through a graph-based objective function to maintain structural consistency. Experimentally, SQ-Diff generates valid and diverse 3D molecules and shows improved performance over existing quantum-based methods. While a gap in generation quality compared to leading classical models remains, our model matches the inference speed of the fastest classical approaches with only a few quantum parameters, advancing the state of the art for pure quantum generative models.
Diffusion models have demonstrated remarkable success in image generation, but they are computationally intensive and time-consuming to train. In this paper, we introduce a novel diffusion model that benefits from quantum computing techniques in order to mitigate computational challenges and enhance generative performance within high energy physics data. The fully quantum diffusion model replaces Gaussian noise with random unitary matrices in the forward process and incorporates a variational quantum circuit within the U-Net in the denoising architecture. We run evaluations on the structurally complex quark and gluon jets dataset from the Large Hadron Collider. The results demonstrate that the fully quantum and hybrid models are competitive with a similar classical model for jet generation, highlighting the potential of using quantum techniques for machine learning problems.
Abstract In this paper, we propose a new methodology to design quantum hybrid diffusion models, derived from classical U-Nets with ResNet and Attention layers. Specifically, we propose two possible different hybridization schemes combining quantum computing’s superior generalization with classical networks’ modularity. In the first one, we acted at the vertex: ResNet convolutional layers are gradually replaced with variational circuits to create Quantum ResNet blocks. In the second proposed architecture, we extend the hybridization to the intermediate level of the encoder, due to its higher sensitivity in the feature extraction process. In order to conduct an in-depth analysis of the potential advantages stemming from the integration of quantum layers, images generated by quantum hybrid diffusion models are compared to those generated by classical models, and evaluated in terms of several quantitative metrics. The results demonstrate an advantage in using hybrid quantum diffusion models, as they generally synthesize better-quality images and converges faster. Moreover, they show the additional advantage of having a lower number of parameters to train compared to the classical one, with a reduction that depends on the extent to which the vertex is hybridized.
Generative models have achieved remarkable success in data synthesis, though recent advances driven by increasing model scale have introduced challenges in computational cost and efficiency. Quantum machine learning offers a promising alternative, representing complex data distributions using compact, highly expressive models. Here, we propose QDiffusion-TS, the first quantum generative diffusion model for time series synthesis, and validate it on the IQM quantum processor. The framework extends a classical diffusion architecture by replacing feed-forward components within the denoising transformer with quantum neural networks, yielding a hybrid quantum transformer that reduces the number of trainable parameters in each replaced component by nearly three orders of magnitude. Evaluated on financial time series from Apple and Amazon, the model generates synthetic data that more accurately reproduces the real distributions, reducing Wasserstein distance by approximately 44% relative to its classical counterpart across both datasets. In a downstream forecasting task, augmentation with the generated data improves predictive performance by up to 71% in RMSE over a baseline trained solely on real data. These results show that quantum enhanced architectures can consistently match and frequently surpass classical performance with substantially fewer parameters, establishing a practical framework towards more efficient and scalable data-driven generative modelling.
Quantum diffusion models are rapidly advancing generative remote-sensing analytics by harnessing quantum computation to deliver higher-quality, more realistic satellite images, while converging faster and with substantially fewer trainable parameters than comparable classical diffusers. In this context, we first realise a fully-quantum latent diffusion model (QLDM) whose denoiser is a variational-quantum circuit acting on a 10-dimensionaI latent code. On EuroSAT, focusing on three land-cover classes Forest, Herbaceous Vegetation and SeaLake, QLDM lowers the Fréchet Inception Distance (FID) and Kernel Inception Distance (KID) by 21.5 % and 29.9 %, respectively, relative to a parameter-matched classical latent diffuser. Yet the severe spatial compression intrinsic to latent processing limits fine-grained detail. To restore image fidelity, we develop a hybrid quantum—classical architecture—the Quanvolutional Conditioned U-Net (QCU-Net)—which inserts entangling quantum layers both at the U-Net bottleneck and as a quanvolutional filter early in the encoder. Trained on the full ten-class EuroSAT RGB set, QCU-Net reaches an FID of 2.57 and a KID of 0.0008, representing 64 % and 76 % improvements over the best classical diffusion baseline (FID = 7.22; KID = 0.0034), and boosts class-conditioning accuracy to 81.7 % versus 62.2 % for the classical model. Thèse gains confirm that embedding quantum circuits within the feature-extraction pipeline yields richer spatial— spectral representations than latent-space quantum processing alone. Our results outline a clear progression in quantum generative modelling for Earth Observation: by first demonstrating with the QLDM that fully quantum denoising is not only feasible but also highly parameter- efficient, then showing how the hybrid QCU-Net architecture restores perceptual fidelity by embedding quantum feature extractors within a classical U-Net backbone, and finally pointing toward a fully pixel-space quantum diffusion model as future work—one that would leverage deep quantum circuits to perform both the forward and reverse diffusion directly in Hilbert space, remove classical auto-encoding steps, and promise even greater image quality and efficiency on fault-tolerant hardware.
Deep generative models are key-enabling technology to computer vision, text generation, and large language models. Denoising diffusion probabilistic models (DDPMs) have recently gained much attention due to their ability to generate diverse and high-quality samples in many computer vision tasks, as well as to incorporate flexible model architectures and a relatively simple training scheme. Quantum generative models, empowered by entanglement and superposition, have brought new insight to learning classical and quantum data. Inspired by the classical counterpart, we propose the quantum denoising diffusion probabilistic model (QuDDPM) to enable efficiently trainable generative learning of quantum data. QuDDPM adopts sufficient layers of circuits to guarantee expressivity, while it introduces multiple intermediate training tasks as interpolation between the target distribution and noise to avoid barren plateau and guarantee efficient training. We provide bounds on the learning error and demonstrate QuDDPM's capability in learning correlated quantum noise model, quantum many-body phases, and topological structure of quantum data. The results provide a paradigm for versatile and efficient quantum generative learning.
Hybrid artificial intelligence–quantum (AI–Q) models arise at the intersection of two rapidly advancing computational paradigms, promising unprecedented capabilities for real-time molecular structure prediction. In classical computational chemistry, the accurate determination of ground-state energies and optimized geometries relies on ab initio methods such as density functional theory (DFT) and coupled-cluster approaches, which, while highly precise, exhibit steep polynomial or exponential scaling with system size. Conversely, deep neural networks (DNNs) have demonstrated remarkable success in learning potential energy surfaces (PES) from large datasets, achieving rapid inference but sometimes lacking the ultimate chemical accuracy required for predictive applications in drug discovery or materials design. On the other hand, variational quantum eigensolvers (VQEs) leverage the intrinsic parallelism of quantum hardware to approximate electronic structure problems, yet they remain constrained by qubit coherence times, gate fidelities, and circuit depth limitations inherent to noisy intermediate-scale quantum (NISQ) devices.