cs.LGNov 14, 2026

Test-Time Constraint-Aware Diffusion Predictive Control

Authors: Braghadeesh Lakshminarayanan, Ojas Phadake, Qi Chen, Dominik Baumann

Organizations: ELLIS Institute Finland · Also part of Cyber-physical Systems Group, Aalto University, Espoo, Finland. · Indian Institute of Technology, Madras, India · ELLIS Institute Finland & Aalto University, Espoo, Finland · 

Abstract

Diffusion models can learn multimodal distributions over finite-horizon state-input sequences, providing a way to generate candidates for predictive control. However, sequences generated by a pretrained model may violate safety constraints introduced at deployment, for example when obstacles are added or repositioned. Retraining whenever these constraints change limits the usefulness of learned predictive controllers. In this paper, we propose Test-Time Constraint-Aware Diffusion Predictive Control (TCA-DPC), an algorithm for adapting the sampling distribution of a pretrained diffusion controller to newly introduced safety constraints without retraining. We treat the pretrained model as a prior over finite-horizon sequences and define an adapted distribution through exponential tilting based on a constraint-violation penalty. The tilt strength is adjusted online using the fraction of generated sequences that fail constraint verification. We approximately sample the adapted distribution using the Metropolis-adjusted Langevin algorithm in the Gaussian noise space of the pretrained diffusion sampler.

Figures & tables

Appendix figures & tables3 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. P3: Persistent Particle Planning for Constrained Diffusion Control

    Oct 5, 2026Hikmet Simsir, Mahyar Fardinfar, Ozgur S. OguzConstrained Motion PlanningTrajectory Generation

  2. From Uncertain to Safe: Conformal Adaptation of Diffusion Models for Safe PDE Control

    Feb 4, 2025Peiyan Hu, Xiaowei Qian, Wenhao Deng +8Diffusion Model GuidanceSafety-Critical Control