Control Consistency Losses for Diffusion Bridges

arXiv:2512.05070v2 Announce Type: replace-cross Abstract: Simulating the conditioned dynamics of diffusion processes, given their initial and terminal states, is an important but challenging problem in the sciences. The difficulty is particularly pronounced for rare events, for which the unconditioned dynamics rarely reach the terminal state. In this work, we propose a novel approach for learning diffusion bridges based on a self-consistency property of the optimal control. The resulting algorithm learns the conditioned dynamics in an iterative online manner, and exhibits strong performance in a range of empirical settings without requiring differentiation through simulated trajectories. Beyond the diffusion bridge setting, we draw connections between our self-consistency framework and recent advances in the wider stochastic optimal control literature.

Related

• Diffusion Sequence Models for Generative In-Context Meta-Learning of Robot Dynamics
• Real-Time Streamable Generative Speech Restoration with Flow Matching
• Training-Free Multi-User Generative Semantic Communications via Null-Space Diffusion Sampling
• Projected Coupled Diffusion for Test-Time Constrained Joint Generation

Source: arXiv cs.LG | 2026-04-23