Leveraging Quantum Measurement Randomness for Shot-Efficient Diffusion Models
Abstract
In quantum computing, the readout always carries finite-shot sampling noise. Suppressing this noise through repeated measurements raises the cost of a quantum neural network (QNN). Meanwhile, a stochastic diffusion sampler deliberately adds randomness at each reverse (generation) step. This paper turns the measurement noise from a liability into a resource. Specifically, we use the noise quantum measurement produces anyway to replace part of the noise the sampler would otherwise inject. We characterize the measurement error as multinomial sampling noise propagated through a trainable linear readout. To connect the circuit's Born probabilities to the denoising task, we further formulate the diffusion process on a bounded zero-sum hyperplane mapped to the probability simplex. Therefore, we can employ a compensation noise to match the mean and covariance of the shot-noise-free linear update under some minimum shot conditions. Solving this condition for the shot count gives a state-dependent adaptive budget. For the denoising diffusion implicit model (DDIM), this budget increases as the sampler becomes more deterministic. On an eight-dimensional Gaussian mixture, the schedule achieves comparable generation quality with $303$ shots per sample against a high-shot reference of 1.6×106, about 5300× fewer. Latent diffusion experiments on MNIST further validate the proposed mechanism for generation in a learned latent space.
Related
- Leveraging verifiable structure in quantum circuits to find more efficient compilations with agentic AI
- Measurement-Efficient Differentiable Quantum Architecture Search for Combinatorial Optimization
- Efficient measurement schemes for the Monte Carlo projective quantum eigensolver
- Measurement-efficient energy estimator for quantum chemistry of photodynamic materials
- MBUndo: Sound, Automatic, and Efficient Measurement-Based Uncomputation for Quantum Circuits
BibTeX
@misc{guo2026leveraging,
title = {{Leveraging Quantum Measurement Randomness for Shot-Efficient Diffusion Models}},
author = {Tianhao Guo and Dianxin Zheng and Yuhua Qian and Yaohui Zheng and Yuzhi Yang and Merouane Debbah and Chau Yuen},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.41721},
eprint = {2610.41721},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.41721},
}
RIS
TY - UNPB
TI - Leveraging Quantum Measurement Randomness for Shot-Efficient Diffusion Models
AU - Tianhao Guo
AU - Dianxin Zheng
AU - Yuhua Qian
AU - Yaohui Zheng
AU - Yuzhi Yang
AU - Merouane Debbah
AU - Chau Yuen
PY - 2026
DA - 2026/10/30
PB - arXiv
DO - 10.48550/arxiv.2610.41721
UR - https://arxiv.org/abs/2610.41721
AB - In quantum computing, the readout always carries finite-shot sampling noise. Suppressing this noise through repeated measurements raises the cost of a quantum neural network (QNN). Meanwhile, a stochastic diffusion sampler deliberately adds randomness at each reverse (generation) step. This paper turns the measurement noise from a liability into a resource. Specifically, we use the noise quantum measurement produces anyway to replace part of the noise the sampler would otherwise inject. We characterize the measurement error as multinomial sampling noise propagated through a trainable linear readout. To connect the circuit's Born probabilities to the denoising task, we further formulate the diffusion process on a bounded zero-sum hyperplane mapped to the probability simplex. Therefore, we can employ a compensation noise to match the mean and covariance of the shot-noise-free linear update under some minimum shot conditions. Solving this condition for the shot count gives a state-dependent adaptive budget. For the denoising diffusion implicit model (DDIM), this budget increases as the sampler becomes more deterministic. On an eight-dimensional Gaussian mixture, the schedule achieves comparable generation quality with $303$ shots per sample against a high-shot reference of 1.6×10^6, about 5300× fewer. Latent diffusion experiments on MNIST further validate the proposed mechanism for generation in a learned latent space.
ER -