paper-with-me

홈 › Papers

Improving Flow Matching by Aligning Flow Divergence

2026-01-31 · Yuhao Huang, Taos Transue, Shih-Hsin Wang, William Feldman, Hong Zhang, Bao Wang arxiv

Conditional flow matching (CFM) stands out as an efficient, simulation-free approach for training flow-based generative models, achieving remarkable performance for data generation. However, CFM is insufficient to ensure accuracy in learning probability paths. In this paper, we introduce a new partial differential equation characterization for the error between the learned and exact probability paths, along with its solution. We show that the total variation gap between the two probability paths is bounded above by a combination of the CFM loss and an associated divergence loss. This theoretical insight leads to the design of a new objective function that simultaneously matches the flow and its divergence. Our new approach improves the performance of the flow-based generative model by a noticeable margin without sacrificing generation efficiency. We showcase the advantages of this enhanced training approach over CFM on several important benchmark tasks, including generative modeling for dynamical systems, DNA sequences, and videos. Code is available at \href{https://github.com/Utah-Math-Data-Science/Flow_Div_Matching}{Utah-Math-Data-Science}.

📄 PDF Abstract BibTeX arXiv:2602.00869

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

On Flow Matching KL Divergence

2025-11-07 · Maojiang Su, Jerry Yao-Chieh Hu, Sophia Pi, Han Liu arxiv

We derive a deterministic, non-asymptotic upper bound on the Kullback-Leibler (KL) divergence of the flow-matching distribution approximation. In particular, if the $L_2$ flow-matching loss is bounded by $ε^2 > 0$, then …

Flow-DPPO: Divergence Proximal Policy Optimization for Flow Matching Models

2026-06-09 · Bowen Ping, Xiangxin Zhou, Penghui Qi, Minnan Luo 외 arxiv

Recent work has demonstrated that online reinforcement learning (RL) can substantially improve the quality and alignment of flow matching models for image and video generation. Methods such as Flow-GRPO and CPS cast the …

Reinforcement LearningVideo Generation

FAIL: Flow Matching Adversarial Imitation Learning for Image Generation

2026-02-12 · Yeyao Ma, Chen Li, Xiaosong Zhang, Han Hu 외 arxiv

Post-training of flow matching models-aligning the output distribution with a high-quality target-is mathematically equivalent to imitation learning. While Supervised Fine-Tuning mimics expert demonstrations effectively,…

Video GenerationImage Generation

Time dependent loss reweighting for flow matching and diffusion models is theoretically justified

2025-11-20 · Lukas Billera, Hedwig Nora Nordlinder, Ben Murrell arxiv

This brief note clarifies that, in Generator Matching (which subsumes a large family of flow matching and diffusion models over continuous, manifold, and discrete spaces), both the Bregman divergence loss and the linear …

Distribution Matching Distillation without Fake Score Network

2026-05-19 · Youngjoong Kim, Deokyeong Lee, Jaesik Park arxiv

Distribution Matching Distillation (DMD) provides an effective distribution-level correction for few-step generation, while relying on an auxiliary fake-score network to track the evolving generative distribution. Recent…