AI RESEARCH
Generative Modeling with Orbit-Space Particle Flow Matching
arXiv CS.CV
•
ArXi:2605.02222v1 Announce Type: cross We present Orbit-Space Geometric Probability Paths (OGPP), a particle-native flow-matching framework for generative modeling of particle systems. OGPP is motivated by two insights: (i) particles are defined up to permutation symmetries, so anonymous indexing inflates per-index target variance and yields curved, hard-to-learn flows; and (ii) particles live in physical space, so the flow terminal velocity has physical meaning and can encode geometric attributes, e.g., surface normals.