AI RESEARCH
SubFlow: Sub-mode Conditioned Flow Matching for Diverse One-Step Generation
arXiv CS.LG
•
ArXi:2604.12273v1 Announce Type: new Flow matching has emerged as a powerful generative framework, with recent few-step methods achieving remarkable inference acceleration. However, we identify a critical yet overlooked limitation: these models suffer from severe diversity degradation, concentrating samples on dominant modes while neglecting rare but valid variations of the target distribution.