A Lightweight Velocity Correction Framework for Improving Instruction Following in Few-Step Text-to-Image Generation

作者：Yating Gao, Zhenlong Zhu, Wen Zihao, Lenan Zhang, Jiexiong Wang, Yang Yang, Guanjun Jiang, Yixiong Zou, Takuya Maekawa · 单位：The University of Osaka, Peking University, Huazhong University of Science and Technology · 会议/期刊：arXiv preprint · 方向：cs.CV · 发布日期：2026-10-12