Self-Supervised Flow Matching for Scalable Multi-Modal Synthesis

ArXi:2603.06507v1 Announce Type: new Strong semantic representations improve the convergence and generation quality of diffusion and flow models. Existing approaches largely rely on external models, which require separate