🤖 AI Summary
This study investigates the efficacy of variational quantum circuits—such as EfficientSU2—in diffusion generative models, with a focus on the failure mechanism of angle embedding under unbounded score-matching objectives due to phase aliasing. To enable a fair evaluation, the authors propose a protocol that integrates quantum modules into DDPM and latent diffusion frameworks via squeeze-and-excitation structures, and conduct systematic assessments using NCSN-based score models, FID metrics, and multiple subsampling significance tests. Experiments show that quantum modules achieve average FID scores comparable to classical baselines despite using 4.5–9 times fewer parameters. Furthermore, applying a π·tanh(·) bounded input transformation effectively mitigates phase aliasing and significantly improves performance; however, the anticipated parameter efficiency advantage of quantum circuits is not realized.
📝 Abstract
We study the integration of variational quantum circuits (VQCs) into diffusion models through a squeeze-and-excitation (SE) channel-modulation scaffold that isolates the quantum contribution. Using a role-matched classical control and multi-seed significance testing across DDPM and latent diffusion on MNIST and CIFAR-10, with a score-based NCSN study on MNIST, we find that quantum cores achieve comparable mean FID to the classical control across DDPM and latent diffusion, while paired sampling-seed tests for EfficientSU2 detect no statistically significant difference. Although the quantum cores use $4.5$--$9\times$ fewer core parameters than the role-matched control, parameter-matched classical controls attain comparable mean FID, so the experiments do not establish a quantum parameter-efficiency advantage. We further identify a structural failure in score-based NCSN: the unbounded score target, proportional to $1/σ$, drives angle-embedding inputs far beyond the $2π$ period of rotation gates, causing phase aliasing and collapse of the quantum modulator. A bounding transformation, $θ\leftarrow π\tanh(\cdot)$, maps inputs to the non-aliasing domain and substantially improves both quantum cores. Since all circuits are classically simulated at a few-qubit scale, we do not claim quantum advantage. Instead, the study provides a fair-comparison protocol for quantum-enhanced generative models and a mechanistic account of when and why angle embeddings fail.