Mar 2026· arXiv.org· Vol abs/2603.26783· 0 citations· 37 references
Computer Science
Abstract
Diffusion models can be challenged in the low signal-to-noise regime, where they have to make pixel-level predictions despite the presence of high noise. The geometric intuition is akin to using the finest stroke for oil painting throughout, which may be ineffective. We therefore study \emph{stroke-size control} as a controlled intervention that changes the roughness of the supervised target, predictions and perturbations across timesteps, in an attempt to ease the low signal-to-noise challenge via the prediction target simplification.
This work proposes an adaptive diffusion controller that dynamically adjusts the number of steps to generate high-quality images efficiently, without additional model training, by leveraging a mixture of step schedules with varying step sizes.
Kuluhan Binici, Cihan Acar, Shivam Aggarwal et al.· International Conference on...· 0 citations
Text-to-image latent diffusion models produce unprecedented visual fidelity but remain severely bottlenecked by the computational latency of iterative sampling. While optimizing the discretization of the continuous-time variable offers a powerful, training-free acceleration pathway, the comparative tradeoffs of foundat...
Tan-Yan Bao, Huy-Tan Thai· International Conference on...· 0 citations
Observation Operator Diffusion is proposed, a unified framework that aligns both the supervision trajectory and feature refinement with the intrinsic recovery order of image structures and introduces GL-CoDA, a decoder that injects scale-specific Gaussian-Lanczos observations across decoding stages for coarse-to-fine f...
Shaojie Guo, Li-Chen Ma, Haoyang Tong et al.· 1 citation
The diffusion model is currently the most common method for generating high-fidelity images, but it requires a large amount of data for training and has a high computational cost. Even though ongoing improvements to architecture have made generation more efficient, it is still important to look into the operational bot...
This work observes an interesting phenomenon: the relationship between the change of generation outputs and the scale of initial noise perturbation is highly linear through the diffusion ODE sampling, and proposes a novel C ontrollable and C onstrained S ampling ( CCS) method, along with a new controller algorithm for...
Bo-Wen Song, Ze-Cheng Zhang, Zhaoxu Luo et al.· 0 citations
A diffusion model starts from pure noise and removes it step by step. Somewhere along the way it stops being able to become"anything"and becomes committed to, say, a horse rather than a truck. We measure when this happens, and what a trained model gets wrong about it, on CIFAR-10. The most direct measurement is to free...
Known for his clear and elegant writing style, Bertsekas shaped fields from control and optimization to large-scale computation and artificial intelligence.
