AI RESEARCH
Warm-Start Flow Matching for Guaranteed Fast Text/Image Generation
arXiv CS.LG
•
ArXi:2603.19360v1 Announce Type: new Current auto-regressive (AR) LLMs, diffusion-based text/image generative models, and recent flow matching (FM) algorithms are capable of generating premium quality text/image samples. However, the inference or sample generation in these models is often very time-consuming and computationally demanding, mainly due to large numbers of function evaluations corresponding to the lengths of tokens or the numbers of diffusion steps. This also necessitates heavy GPU resources, time, and electricity.