做生成模型研究的开发者终于有了一个既能高效单步生成、又能按需多步精调的框架——DFM 解决了 Drift 模型缺乏推理扩展性的痛点，值得关注其在不同场景下的灵活应用。
Drift Flow Matching (DFM) 提出了一种新框架，将高效的 Drift 模型（单步生成）与 Flow Matching（多步迭代生成）结合起来。DFM 保留了直接传输映射的效率，同时允许通过多步推理来优化生成质量，从而在计算成本与生成质量之间灵活权衡。实验表明，DFM 在多个任务和数据集上均有效，为生成模型提供了新的自适应采样范式。
Drift Flow Matching
Iterative generative models such as Flow Matching and Diffusion models have demonstrated strong test-time scaling behavior, where additional inference computation can improve generation quality. In contrast, Drift Models offer efficient one-step generation, but their direct generation paradigm limits such flexibility. In this work, we propose Drift Flow Matching (DFM), a framework that connects drifting generative modeling with flow-based iterative generation. DFM preserves the efficiency of direct transport maps while enabling generation to be refined through multiple inference steps when desired. This bridges the gap between one-step Drift Models and multi-step Flow Matching methods, and provides a novel generative paradigm that can adapt sampling computation to different quality--efficiency requirements. Extensive experiments across different tasks and datasets demonstrate the effectiveness and generality of the proposed framework.