用最大似然重构优化流匹配模型，提升生成质量与稳定性。
Fine-Tuning Flow Matching via Maximum Likelihood Estimation of Reconstructions
- 基于重构的最大似然细调，利用平滑微分方程实现无模拟训练
- 在气象预测与机器人控制中显著提升生成性能与鲁棒性
- 支持灵活扩展，适用于需要高稳定性的生成任务
流匹配（Flow Matching, FM）模型在生成任务中表现优异。基于扩散模型，FM采用无模拟训练范式，兼具简洁与高效，但存在训练-推理差距：训练时无法评估模型输出。此外，直线流假设存在固有局限。为此，我们提出通过重构的最大似然估计（MLE）对FM进行细调——得益于FM的光滑常微分方程（ODE）形式，区别于扩散模型中的随机微分方程（SDE）。我们首先在数值精度约束下理论分析了训练损失与推理误差的关系。随后提出一种易于实现的细调框架，支持复杂扩展。在此基础上，引入广义人工黏性项以增强流的稳定性与鲁棒性，并提供直接参数化方法及严格的理论保证。实验表明，该方法在多种场景下有效：小型案例揭示了细调机制，大规模评估在气象预报与机器人操控策略中均验证了可靠的性能提升。
原文摘要 · Abstract (English)
Flow Matching (FM) models achieve remarkable results in generative tasks. Building upon diffusion models, FM's simulation-free training paradigm enables simplicity and efficiency but introduces a train-inference gap: model outputs cannot be assessed during training. Moreover, the straight flow assumption suffers from some inherent limitations. To address this, we propose to fine-tune FM via Maximum Likelihood Estimation (MLE) of reconstructions -- enabled by FM's smooth ODE formulation, unlike the stochastic differential equations (SDEs) in diffusion models. We first theoretically analyze the relationship between training loss and inference error in FM under numerical precision constraints. We then propose an easy-to-implement fine-tuning framework based on MLE of reconstructions, with flexibility for sophisticated extensions. Building on this, we incorporate a generalized artificial viscosity term that enhances flow stability and robustness, accompanied by a direct parameterization method and rigorous theoretical guarantees. Experiments demonstrate our method's effectiveness across diverse settings: a toy example provides mechanistic insights into the fine-tuning process, while large-scale evaluations on meteorological forecasting and robotic manipulation policies validate reliable performance improvements.
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。