Controllable and Constrained Sampling in Diffusion Models via Initial Noise Perturbation

Bo-Wen Song Ze-Cheng Zhang Zhaoxu Luo Jason Hu Wei Yuan Jing Jia Zhengxu Tang Guanyang Wang Li-Yue Shen TraceRoot.AI

· 0 citations · 45 references

TL;DR

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 diffusion models that enables control over both the proximity of individual samples to a target image and the alignment of the sample mean with the target, while preserving good sample quality.

Semantic Scholar

View source

Similar papers

#machine learning Preprint Oct 2026

Adaptive Second-Order Solvers for Fast Stochastic Diffusion Sampling

Diffusion models rely on numerical solvers requiring time-discretization, which has a large influence on the tradeoff between sampling cost and quality. However, the computational difficulty of the reverse process varies along the sampling trajectory and across data distributions, making the choice of discretization im...

E. Kemperman, Luca Ambrogioni · 0 citations

#machine learning Preprint Oct 2026

Wasserstein Gradient Flows and Forward-Only Diffusion Are Not Enough for Multimodal Sampling

There has been a proliferation of sampling algorithms based on Wasserstein gradient flows (WGF) and forward-only diffusion processes (FODP), often accompanied by theoretical guarantees of exponentially fast convergence to the target distribution. These guarantees are frequently interpreted as evidence that such methods...

Daniel McBride, Pratik Khandagale, Cristina Garcia-Cardona et al. · 0 citations

#data science Preprint Sep 2026

FluxLite: Inference-Time Proposal Control for Discrete Diffusion Models

FluxLite is introduced, a lightweight, training-free proposal-control framework for discrete diffusion, identifying a tilted-path coverage factor that governs robustness to score error, together with finite-particle convergence for a fixed controlled Feynman-Kac recursion.

Yinuo Ren, Haoxuan Chen, Grant M. Rotskoff et al. · 1 citation

Preprint Aug 2026

Self-Augmented Diffusion Guidance for Physics-Informed Generation

This study proposes a simple yet effective physics-informed approach based on diffusion guidance with self-generated data augmentation that significantly reduces the deviations compared with standard diffusion models but also achieves further reductions when combined with existing physics-constrained diffusion methods.

Akira Osaka, Naoya Takeishi, T. Yairi · 0 citations

#machine learning Preprint Sep 2026

A Study of Conditional Diffusion Models for Open-Loop Control under Dry Friction and Stiction

Results indicate that conditional diffusion provides an effective mechanism for generating temporally coherent control sequences that overcome stiction by conditioning and recombining structured control primitives from the training prior for state-to-state open-loop control.

Eric A. Antonelo · 0 citations

#machine learning Preprint Oct 2026

Cross-Time Directional Selection in Diffusion Sampling

How strongly do the remaining diffusion-sampling steps amplify a perturbation at a late latent state? Standard measurements answer this question with newly sampled isotropic noise, even though perturbations encountered during sampling have already been transformed by earlier steps. We compare these two cases directly....

Dhia Naouali · 0 citations