FlowSteer：以真实轨迹引导少步图像合成
FlowSteer: Guiding Few-Step Image Synthesis with Authentic Trajectories
把 ReFlow 蒸馏从「点对点回归」改成沿教师真实 ODE 轨迹对齐，并给出 FlowMatchEulerDiscreteScheduler 的少步推理缺陷修复，可直接用于 SD3 类流模型的少步加速训练与采样器实现。
原文依据 guiding the student along teacher's authentic generation trajectories
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论随着流匹配在视觉生成中取得成功，采样效率仍是其实际应用的关键瓶颈。在流模型的加速方法中，ReFlow 尽管与流匹配具有理论一致性，却在某种程度上被忽视，这主要是由于它在实际场景中的表现不如一致性蒸馏和分数蒸馏。本文在 ReFlow 框架内研究该问题，并提出 FlowSteer，一种通过沿教师的真实生成轨迹引导学生来释放基于 ReFlow 的蒸馏潜力的方法。我们首先指出，分段 ReFlow 的性能受训练过程中严重的分布不匹配所制约，并提出在线轨迹对齐（OTA）来解决该问题。随后，我们引入直接作用于 ODE 轨迹的对抗蒸馏目标，提升学生对教师生成轨迹的遵循程度。此外，我们发现并修复了广泛使用的 FlowMatchEulerDiscreteScheduler 中一个此前未被发现的缺陷，该缺陷会大幅降低少步推理质量。在 SD3 上的实验结果证明了本方法的有效性。
英文原文摘要 1220 字符
With the success of flow matching in visual generation, sampling efficiency remains a critical bottleneck for its practical application. Among flow models' accelerating methods, ReFlow has been somehow overlooked although it has theoretical consistency with flow matching. This is primarily due to its suboptimal performance in practical scenarios compared to consistency distillation and score distillation. In this work, we investigate this issue within the ReFlow framework and propose FlowSteer, a method unlocks the potential of ReFlow-based distillation by guiding the student along teacher's authentic generation trajectories. We first identify that Piecewised ReFlow's performance is hampered by a critical distribution mismatch during the training and propose Online Trajectory Alignment(OTA) to resolve it. Then, we introduce a adversarial distillation objective applied directly on the ODE trajectory, improving the student's adherence to the teacher's generation trajectory. Furthermore, we find and fix a previously undiscovered flaw in the widely-used \texttt FlowMatchEulerDiscreteScheduler that largely degrades few-step inference quality. Our experiment result on SD3 demonstrates our method's efficacy.