AI RESEARCH
Extending One-Step Image Generation from Class Labels to Text via Discriminative Text Representation
arXiv CS.CV
•
ArXi:2604.18168v1 Announce Type: new Few-step generation has been a long-standing goal, with recent one-step generation methods exemplified by MeanFlow achieving remarkable results. Existing research on MeanFlow primarily focuses on class-to-image generation. However, an intuitive yet unexplored direction is to extend the condition from fixed class labels to flexible text inputs, enabling richer content creation.