Empirical Scaling Laws of Quantum Vision Transformers
Abstract
Quantum machine learning (QML) is a promising application of quantum computing, yet when and how practical learning models can achieve quantum advantage remains poorly understood. Even a polynomial improvement in the resources needed to attain a given predictive performance would be significant. In this work, we investigate empirical scaling laws of quantum vision transformers (QViTs), which connect QML to the transformer architecture underlying modern large language models. We ask whether QViTs exhibit scaling behavior analogous to classical models and how this behavior changes when qubit number is included alongside dataset size and model parameters. The validation losses in our numerical image-classification experiments are well described by a joint shifted power-law model in qubit number, transformer depth, and training dataset size. Expressing this fit in terms of the classical encoder parameter budget identifies a regime in which increasing the number of qubits can compensate for fewer classical encoder parameters at comparable predicted performance. This finding is consistent with earlier observations of parameter efficiency in quantum learning models and suggests a direction for exploring possible resource savings in QML.
Related
- Scaling Laws of Quantum Networks: An Entanglement Transport Framework
- Crossing over universal scaling laws in two-dimensional driven dissipative condensates
- Thermal scaling laws for open-water swimming
- Dropout Universality: Scaling Laws and Optimal Scheduling at the Edge-of-Chaos
- Detectability limits of scaling laws
BibTeX
@misc{li2026empirical,
title = {{Empirical Scaling Laws of Quantum Vision Transformers}},
author = {Yuqing Li and Fengyi Gao and Xin Jin and Xiaowei Jia and Qi Cheng and Junyu Liu},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.13970},
eprint = {2610.13970},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.13970},
}
RIS
TY - UNPB
TI - Empirical Scaling Laws of Quantum Vision Transformers
AU - Yuqing Li
AU - Fengyi Gao
AU - Xin Jin
AU - Xiaowei Jia
AU - Qi Cheng
AU - Junyu Liu
PY - 2026
DA - 2026/10/01
PB - arXiv
DO - 10.48550/arxiv.2610.13970
UR - https://arxiv.org/abs/2610.13970
AB - Quantum machine learning (QML) is a promising application of quantum computing, yet when and how practical learning models can achieve quantum advantage remains poorly understood. Even a polynomial improvement in the resources needed to attain a given predictive performance would be significant. In this work, we investigate empirical scaling laws of quantum vision transformers (QViTs), which connect QML to the transformer architecture underlying modern large language models. We ask whether QViTs exhibit scaling behavior analogous to classical models and how this behavior changes when qubit number is included alongside dataset size and model parameters. The validation losses in our numerical image-classification experiments are well described by a joint shifted power-law model in qubit number, transformer depth, and training dataset size. Expressing this fit in terms of the classical encoder parameter budget identifies a regime in which increasing the number of qubits can compensate for fewer classical encoder parameters at comparable predicted performance. This finding is consistent with earlier observations of parameter efficiency in quantum learning models and suggests a direction for exploring possible resource savings in QML.
ER -