Attraction, Repulsion, and Friction: Introducing DMF, a Friction-Augmented Drifting Model

arXiv:2604.18194v1 Announce Type: cross Abstract: Drifting Models [Deng et al., 2026] train a one-step generator by evolving samples under a kernel-based drift field, avoiding ODE integration at inference. The original analysis leaves two questions open. The drift-field iteration admits a locally repulsive regime in a two-particle surrogate, and vanishing of the drift (V_{p,q}equiv 0) is not known to force the learned distribution q to match the target p. We derive a contraction threshold for the surrogate and show that a linearly-scheduled friction coefficient gives a finite-horizon bound on the error trajectory. Under a Gaussian kernel we prove that the drift-field equilibrium is identifiable: vanishing of V_{p,q} on any open set forces q=p, closing the converse of Proposition 3.1 of Deng et al. Our friction-augmented model, DMF (Drifting Model with Friction), matches or exceeds Optimal Flow Matching on FFHQ adult-to-child domain translation at 16x lower training compute.

Related

• Noise-Adaptive Diffusion Sampling for Inverse Problems Without Task-Specific Tuning
• Time-reversed Flow Matching with Worst Transport in High-dimensional Latent Space for Image Anomaly Detection
• A Single Image and Multimodality Is All You Need for Novel View Synthesis
• DiffusionPrint: Learning Generative Fingerprints for Diffusion-Based Inpainting Localization

Source: arXiv cs.CV | 2026-04-21