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:2512.01519

QuantumCanvas: A Multimodal Benchmark for Visual Learning of Atomic Interactions

Can Polat, Erchin Serpedin, Mustafa Kurban, Hasan Kurban

cs.CVcond-mat.mtrl-sciquant-phSubmitted Dec 1, 2025• Announced Dec 2, 2025
arXivHTMLPDF

Abstract

Despite rapid advances in molecular and materials machine learning, most models still lack physical transferability: they fit correlations across whole molecules or crystals rather than learning the quantum interactions between atomic pairs. Yet bonding, charge redistribution, orbital hybridization, and electronic coupling all emerge from these two-body interactions that define local quantum fields in many-body systems. We introduce QuantumCanvas, a large-scale multimodal benchmark that treats two-body quantum systems as foundational units of matter. The dataset spans 2,850 element-element pairs, each annotated with 18 electronic, thermodynamic, and geometric properties and paired with ten-channel image representations derived from l- and m-resolved orbital densities, angular field transforms, co-occupancy maps, and charge-density projections. These physically grounded images encode spatial, angular, and electrostatic symmetries without explicit coordinates, providing an interpretable visual modality for quantum learning. Benchmarking eight architectures across 18 targets, we report mean absolute errors of 0.201 eV on energy gap using GATv2, 0.265 eV on HOMO and 0.274 eV on LUMO using EGNN. For energy-related quantities, DimeNet attains 2.27 eV total-energy MAE and 0.132 eV repulsive-energy MAE, while a multimodal fusion model achieves a 2.15 eV Mermin free-energy MAE. Pretraining on QuantumCanvas further improves convergence stability and generalization when fine-tuned on larger datasets such as QM9, MD17, and CrysMTM. By unifying orbital physics with vision-based representation learning, QuantumCanvas provides a principled and interpretable basis for learning transferable quantum interactions through coupled visual and numerical modalities. Dataset and model implementations are available at https://github.com/KurbanIntelligenceLab/QuantumCanvas.

Related papers

Predicting Molecular Properties With Quantum Kernels: a Study on the Qm9 Dataset

Sonam Khattar, Harshit Kumawat, Aryan

Predicting molecular properties is a cornerstone of drug discovery, materials science, and quantum chemistry, where accuracy and speed are critical. Classical machine learning mod- els offer rapid predictions but often fail to capture the intricate quantum interactions governing molecular behavior. Our team investigated quantum kernel methods to predict properties of the QM9 dataset, a widely recognized benchmark containing$\mathbf{1 3 4, 0 0 0}$organic molecules with diverse chemical structures. By encoding molecular features into quantum circuits, we mapped classical data into highdimensional Hilbert spaces to enhance prediction accuracy. Our pipeline encompasses data preprocessing, quantum circuit design, model training with support vector regression, and evaluation of ten key molecular properties. The quantum kernel approach delivered promising results, achieving mean absolute errors of 0.13 Debye for dipole moment and 0.08 eV for HOMOLUMO gap, even when executed on noisy quantum devices. This work provides a comprehensive, reproducible pipeline, benchmarks quantum and classical models, and discusses practical challenges, contributing significantly to the field of quantumenhanced computational chemistry.

IEEE International Conference for Women in Innovation, Technology & Entrepreneurship (ICWITE)•Sep 26, 2025•0 citations

Towards Quantum-Driven Multimodal Machine Learning: Methods, Challenges, and Future Directions

Debashis Das, Jaclyn Claiborne, Lexus Brinkley-Tapp +3 more

Artificial intelligence (AI) has achieved remarkable progress in single-modal tasks such as text analysis and image recognition. Several real-world problems often demand multimodal intelligence that can integrate and reason across diverse data sources. Traditional machine learning methods face scalability and efficiency limitations for handling heterogeneous multimodal datasets. Recent work in quantum computing introduces new opportunities for AI using superposition and entanglement. Quantum Machine Learning (QML) has shown early promise in classification, optimization, and learning, but applications to multimodal intelligence remain at an early stage. Existing studies are fragmented, and there is a lack of systematic understanding of how QML can address the challenges of multimodal learning. Therefore, this paper provides a structured survey of QML approaches for multimodal tasks. We review data encoding strategies, quantum models such as quantum support vector machines and variational quantum circuits, and hybrid quantum-classical frameworks. We analyze their strengths, limitations, current applications, and future directions. Our findings show that QML faces obstacles in scalability, noise, and benchmarking for efficient cross-modal reasoning and representation. Quantum generative models and quantum-inspired multimodal architectures could significantly expand the capabilities of AI in the future. It highlights both the opportunities and open challenges in applying QML to multimodal intelligence. Overall, the paper provides researchers with a roadmap for advancing this field and sets the stage for future applications in precision medicine, autonomous systems, cybersecurity, and climate modeling.

IEEE International Conference on Trust, Privacy and Security in Intelligent Systems and Applications (TPS-ISA)•Nov 12, 2025•0 citations

QCalEval: Benchmarking Vision-Language Models for Quantum Calibration Plot Understanding

Shuxiang Cao, Zijian Zhang, Abhishek Agarwal +29 more

Quantum computing calibration depends on interpreting experimental data, and calibration plots provide the most universal human-readable representation for this task, yet no systematic evaluation exists of how well vision-language models (VLMs) interpret them. We introduce QCalEval, the first VLM benchmark for quantum calibration plots: 243 samples across 87 scenario types from 22 experiment families, spanning superconducting qubits and neutral atoms, evaluated on six question types in both zero-shot and in-context learning settings. The best general-purpose zero-shot model reaches a mean score of 72.3, and many open-weight models degrade under multi-image in-context learning, whereas frontier closed models improve substantially. A supervised fine-tuning ablation at the 9-billion-parameter scale shows that SFT improves zero-shot performance but cannot close the multimodal in-context learning gap. As a reference case study, we release NVIDIA Ising Calibration 1, an open-weight model based on Qwen3.5-35B-A3B that reaches 74.7 zero-shot average score.

quant-pharXiv:2604.25884•Apr 29, 2026•11 citations

Quantum-centric machine learning for molecular dynamics

Yanxian Tao, Lingyun Wan, Xiongzhi Zeng +4 more

Accurate and efficient prediction of electronic wavefunctions is central to ab initio molecular dynamics (AIMD) and electronic structure theory. However, conventional ab initio methods require self-consistent optimization of electronic states at every nuclear configuration, leading to prohibitive computational costs, especially for large or strongly correlated systems. Here, we introduce a quantum-centric machine learning (QCML) model-a hybrid quantum-classical framework that integrates parameterized quantum circuits (PQCs) with Transformer-based machine learning to directly predict molecular wavefunctions and quantum observables. By pretraining the Transformer on a diverse dataset of molecules and ansatz types and subsequently fine-tuning it for specific systems, QCML learns transferable mappings between molecular descriptors and PQC parameters, eliminating the need for iterative variational optimization. The pretrained model achieves chemical accuracy in potential energy surfaces, atomic forces, and dipole moments across multiple molecules and ansatzes, and enables efficient AIMD simulations with infrared spectra prediction. This work establishes a scalable and transferable quantum-centric machine learning paradigm, bridging variational quantum algorithms and modern deep learning for next-generation molecular simulation and quantum chemistry applications.

physics.chem-phCross-listedarXiv:2511.07771•Nov 12, 2025•2 citations

Quantum assistant: Specialization of multimodal vision-language models for quantum computing

Samuel Lima Braz, João Paulo Leite

Code assistants for quantum computing operate exclusively on text, disregarding visual representations (circuit diagrams, Bloch spheres, and histograms) that permeate the domain’s documentation. This work is presented as a resource-and-pilot study with three contributions: (1) the first public multimodal dataset for Qiskit-oriented quantum programming, containing 8,366 samples (45% with images) across function completion, code generation, and question answering tasks, released under Apache 2.0; (2) a reproducible synthetic generation pipeline that extracts content from public Qiskit documentation, transcribes images via a multimodal model, generates question-answer pairs, and validates code through automatically generated unit tests; and (3) a feasibility study of parameter-efficient VLM specialization, in which rsLoRA (rank 32, two epochs) on Qwen3-VL-8B-Instruct is retained as the best configuration under a controlled method-selection comparison of five LoRA variants. On the text-only external benchmarks Qiskit HumanEval and Qiskit HumanEval Hard, which measure quantum-code specialization rather than visual understanding, the fine-tuned model reaches 41.72% and 30.46% Pass@1 respectively, above the IBM-specialized Qwen2.5-Coder-14B-Qiskit (40.40% / 25.83%) and granite-3.3-8b-qiskit (17.88% / 9.93%), and + 17.22 / + 15.89 pp over the Qwen3-VL-8B-Instruct base. Visual capability is assessed on the synthetic multimodal test split, where multimodal samples reach 64.29% Pass@1 versus 46.33% on text-only samples ( + 17.96 pp); a matched training ablation that withholds the multimodal portion of the training set localizes + 6.25 pp of that gap to the multimodal training signal. The dataset, models, and code are available under the Apache 2.0 license.

Expert Systems with Applications•Aug 13, 2026•1 citation

Adapting Quantum Machine Learning for Energy Dissociation of Bonds

Swathi Chandrasekhar, Shiva Raj Pokhrel, Navneet Singh

Accurate prediction of bond dissociation energies (BDEs) underpins mechanistic insight and the rational design of molecules and materials. We present a systematic, reproducible benchmark comparing quantum and classical machine learning models for BDE prediction using a chemically curated feature set encompassing atomic properties (atomic numbers, hybridization), bond characteristics (bond order, type), and local environmental descriptors. Our quantum framework, implemented in Qiskit Aer on six qubits, employs ZZFeatureMap encodings with variational ansätz (RealAmplitudes) across multiple architectures Variational Quantum Regressors (VQR), Quantum Support Vector Regressors (QSVR), Quantum Neural Networks (QNN), Quantum Convolutional Neural Networks (QCNN), and Quantum Random Forests (QRF). These are rigorously benchmarked against strong classical baselines, including Support Vector Regression (SVR), Random Forests (RF), and Multi-Layer Perceptrons (MLP). Comprehensive evaluation spanning absolute and relative error metrics, threshold accuracies, and error distributions shows that top-performing quantum models (QCNN, QRF) match the predictive accuracy and robustness of classical ensembles and deep networks, particularly within the chemically prevalent mid-range BDE regime. These findings establish a transparent baseline for quantum-enhanced molecular property prediction and outline a practical foundation for advancing quantum computational chemistry toward near-chemical accuracy.

arXiv:2510.06563•Sep 25, 2025•4 citations
Despite rapid advances in molecular and materials machine learning, most models still lack physical transferability: they fit correlations across whole molecules or crystals rather than learning the quantum interactions between atomic pairs. Yet bonding, charge redistribution, orbital hybridization, and electronic coupling all emerge from these two-body interactions that define local quantum fields in many-body systems. We introduce QuantumCanvas, a large-scale multimodal benchmark that treats two-body quantum systems as foundational units of matter. The dataset spans 2,850 element-element pairs, each annotated with 18 electronic, thermodynamic, and geometric properties and paired with ten-channel image representations derived from l- and m-resolved orbital densities, angular field transforms, co-occupancy maps, and charge-density projections. These physically grounded images encode spatial, angular, and electrostatic symmetries without explicit coordinates, providing an interpretable visual modality for quantum learning. Benchmarking eight architectures across 18 targets, we report mean absolute errors of 0.201 eV on energy gap using GATv2, 0.265 eV on HOMO and 0.274 eV on LUMO using EGNN. For energy-related quantities, DimeNet attains 2.27 eV total-energy MAE and 0.132 eV repulsive-energy MAE, while a multimodal fusion model achieves a 2.15 eV Mermin free-energy MAE. Pretraining on QuantumCanvas further improves convergence stability and generalization when fine-tuned on larger datasets such as QM9, MD17, and CrysMTM. By unifying orbital physics with vision-based representation learning, QuantumCanvas provides a principled and interpretable basis for learning transferable quantum interactions through coupled visual and numerical modalities. Dataset and model implementations are available at https://github.com/KurbanIntelligenceLab/QuantumCanvas.
Predicting molecular properties is a cornerstone of drug discovery, materials science, and quantum chemistry, where accuracy and speed are critical. Classical machine learning mod- els offer rapid predictions but often fail to capture the intricate quantum interactions governing molecular behavior. Our team investigated quantum kernel methods to predict properties of the QM9 dataset, a widely recognized benchmark containing134,000\mathbf{1 3 4, 0 0 0}134,000organic molecules with diverse chemical structures. By encoding molecular features into quantum circuits, we mapped classical data into highdimensional Hilbert spaces to enhance prediction accuracy. Our pipeline encompasses data preprocessing, quantum circuit design, model training with support vector regression, and evaluation of ten key molecular properties. The quantum kernel approach delivered promising results, achieving mean absolute errors of 0.13 Debye for dipole moment and 0.08 eV for HOMOLUMO gap, even when executed on noisy quantum devices. This work provides a comprehensive, reproducible pipeline, benchmarks quantum and classical models, and discusses practical challenges, contributing significantly to the field of quantumenhanced computational chemistry.
Artificial intelligence (AI) has achieved remarkable progress in single-modal tasks such as text analysis and image recognition. Several real-world problems often demand multimodal intelligence that can integrate and reason across diverse data sources. Traditional machine learning methods face scalability and efficiency limitations for handling heterogeneous multimodal datasets. Recent work in quantum computing introduces new opportunities for AI using superposition and entanglement. Quantum Machine Learning (QML) has shown early promise in classification, optimization, and learning, but applications to multimodal intelligence remain at an early stage. Existing studies are fragmented, and there is a lack of systematic understanding of how QML can address the challenges of multimodal learning. Therefore, this paper provides a structured survey of QML approaches for multimodal tasks. We review data encoding strategies, quantum models such as quantum support vector machines and variational quantum circuits, and hybrid quantum-classical frameworks. We analyze their strengths, limitations, current applications, and future directions. Our findings show that QML faces obstacles in scalability, noise, and benchmarking for efficient cross-modal reasoning and representation. Quantum generative models and quantum-inspired multimodal architectures could significantly expand the capabilities of AI in the future. It highlights both the opportunities and open challenges in applying QML to multimodal intelligence. Overall, the paper provides researchers with a roadmap for advancing this field and sets the stage for future applications in precision medicine, autonomous systems, cybersecurity, and climate modeling.
Quantum computing calibration depends on interpreting experimental data, and calibration plots provide the most universal human-readable representation for this task, yet no systematic evaluation exists of how well vision-language models (VLMs) interpret them. We introduce QCalEval, the first VLM benchmark for quantum calibration plots: 243 samples across 87 scenario types from 22 experiment families, spanning superconducting qubits and neutral atoms, evaluated on six question types in both zero-shot and in-context learning settings. The best general-purpose zero-shot model reaches a mean score of 72.3, and many open-weight models degrade under multi-image in-context learning, whereas frontier closed models improve substantially. A supervised fine-tuning ablation at the 9-billion-parameter scale shows that SFT improves zero-shot performance but cannot close the multimodal in-context learning gap. As a reference case study, we release NVIDIA Ising Calibration 1, an open-weight model based on Qwen3.5-35B-A3B that reaches 74.7 zero-shot average score.
Accurate and efficient prediction of electronic wavefunctions is central to ab initio molecular dynamics (AIMD) and electronic structure theory. However, conventional ab initio methods require self-consistent optimization of electronic states at every nuclear configuration, leading to prohibitive computational costs, especially for large or strongly correlated systems. Here, we introduce a quantum-centric machine learning (QCML) model-a hybrid quantum-classical framework that integrates parameterized quantum circuits (PQCs) with Transformer-based machine learning to directly predict molecular wavefunctions and quantum observables. By pretraining the Transformer on a diverse dataset of molecules and ansatz types and subsequently fine-tuning it for specific systems, QCML learns transferable mappings between molecular descriptors and PQC parameters, eliminating the need for iterative variational optimization. The pretrained model achieves chemical accuracy in potential energy surfaces, atomic forces, and dipole moments across multiple molecules and ansatzes, and enables efficient AIMD simulations with infrared spectra prediction. This work establishes a scalable and transferable quantum-centric machine learning paradigm, bridging variational quantum algorithms and modern deep learning for next-generation molecular simulation and quantum chemistry applications.
Code assistants for quantum computing operate exclusively on text, disregarding visual representations (circuit diagrams, Bloch spheres, and histograms) that permeate the domain’s documentation. This work is presented as a resource-and-pilot study with three contributions: (1) the first public multimodal dataset for Qiskit-oriented quantum programming, containing 8,366 samples (45% with images) across function completion, code generation, and question answering tasks, released under Apache 2.0; (2) a reproducible synthetic generation pipeline that extracts content from public Qiskit documentation, transcribes images via a multimodal model, generates question-answer pairs, and validates code through automatically generated unit tests; and (3) a feasibility study of parameter-efficient VLM specialization, in which rsLoRA (rank 32, two epochs) on Qwen3-VL-8B-Instruct is retained as the best configuration under a controlled method-selection comparison of five LoRA variants. On the text-only external benchmarks Qiskit HumanEval and Qiskit HumanEval Hard, which measure quantum-code specialization rather than visual understanding, the fine-tuned model reaches 41.72% and 30.46% Pass@1 respectively, above the IBM-specialized Qwen2.5-Coder-14B-Qiskit (40.40% / 25.83%) and granite-3.3-8b-qiskit (17.88% / 9.93%), and + 17.22 / + 15.89 pp over the Qwen3-VL-8B-Instruct base. Visual capability is assessed on the synthetic multimodal test split, where multimodal samples reach 64.29% Pass@1 versus 46.33% on text-only samples ( + 17.96 pp); a matched training ablation that withholds the multimodal portion of the training set localizes + 6.25 pp of that gap to the multimodal training signal. The dataset, models, and code are available under the Apache 2.0 license.
Accurate prediction of bond dissociation energies (BDEs) underpins mechanistic insight and the rational design of molecules and materials. We present a systematic, reproducible benchmark comparing quantum and classical machine learning models for BDE prediction using a chemically curated feature set encompassing atomic properties (atomic numbers, hybridization), bond characteristics (bond order, type), and local environmental descriptors. Our quantum framework, implemented in Qiskit Aer on six qubits, employs ZZFeatureMap encodings with variational ansätz (RealAmplitudes) across multiple architectures Variational Quantum Regressors (VQR), Quantum Support Vector Regressors (QSVR), Quantum Neural Networks (QNN), Quantum Convolutional Neural Networks (QCNN), and Quantum Random Forests (QRF). These are rigorously benchmarked against strong classical baselines, including Support Vector Regression (SVR), Random Forests (RF), and Multi-Layer Perceptrons (MLP). Comprehensive evaluation spanning absolute and relative error metrics, threshold accuracies, and error distributions shows that top-performing quantum models (QCNN, QRF) match the predictive accuracy and robustness of classical ensembles and deep networks, particularly within the chemically prevalent mid-range BDE regime. These findings establish a transparent baseline for quantum-enhanced molecular property prediction and outline a practical foundation for advancing quantum computational chemistry toward near-chemical accuracy.