为流匹配模型设计了不改变采样路径的推理期计算扩展方法
Inference-Time Compute Scaling For Flow Matching
- 保留线性插值路径，动态增加推理计算量
- 图像生成与蛋白质生成任务中样本质量随算力提升而持续改善
- 首次将推理算力扩展用于科学领域，适用于蛋白生成等任务
在大语言模型和基于扩散的图像生成中，推理阶段增加计算量已被证明可提升生成质量。与此同时，流匹配（Flow Matching, FM）在语言、视觉和科学领域日益受到关注，但其推理阶段的算力扩展方法仍鲜有研究。现有工作如Kim等人（2025）虽尝试解决该问题，却在推理时改用非线性的方差保持插值路径，牺牲了FM原有的高效直线采样特性。此外，当前的推理算力扩展仅应用于视觉任务，如图像生成。本文提出新的推理期扩展方法，保持采样过程中的线性插值结构不变。在图像生成任务以及首次（据我们所知）在无条件蛋白质生成任务上的评估表明：1）随着推理计算量增加，生成样本质量持续提升；2）流匹配的推理算力扩展可成功拓展至科学领域。
原文摘要 · Abstract (English)
Allocating extra computation at inference time has recently improved sample quality in large language models and diffusion-based image generation. In parallel, Flow Matching (FM) has gained traction in language, vision, and scientific domains, but inference-time scaling methods for it remain under-explored. Concurrently, Kim et al., 2025 approach this problem but replace the linear interpolant with a non-linear variance-preserving (VP) interpolant at inference, sacrificing FM's efficient and straight sampling. Additionally, inference-time compute scaling for flow matching has only been applied to visual tasks, like image generation. We introduce novel inference-time scaling procedures for FM that preserve the linear interpolant during sampling. Evaluations of our method on image generation, and for the first time (to the best of our knowledge), unconditional protein generation, show that I) sample quality consistently improves as inference compute increases, and II) flow matching inference-time scaling can be applied to scientific domains.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。