CAR-Flow: Condition-Aware Reparameterization Aligns Source and Target for Better Flow Matching

Chen Chen, Pengsheng Guo, Liangchen Song, Jiasenlu, Rui Qian, wangxinze, Tsu-Jui Fu, Wei Liu, Yinfei Yang, Alex Schwing

CAR-Flow: Condition-Aware Reparameterization Aligns Source and Target for Better Flow Matching: 7 upvotes on Hugging Face Daily Papers, #15 of 24 papers on 2025-09-24. Day-by-day upvote history.

Conditional generative modeling aims to learn a conditional data distribution from samples containing data-condition pairs. For this, diffusion and flow-based methods have attained compelling results. These methods use a learned (flow) model to transport an initial standard Gaussian noise that ignores the condition to the conditional data distribution. The model is hence required to learn both mass transport and conditional injection. To ease the demand on the model, we propose Condition-Aware Reparameterization for Flow Matching (CAR-Flow) -- a lightweight, learned shift that conditions the source, the target, or both distributions. By relocating these distributions, CAR-Flow shortens the probability path the model must learn, leading to faster training in practice. On low-dimensional synthetic data, we visualize and quantify the effects of CAR. On higher-dimensional natural image data (ImageNet-256), equipping SiT-XL/2 with CAR-Flow reduces FID from 2.07 to 1.68, while introducing less than 0.6% additional parameters.

Paper page on Hugging Face · arXiv

Data: hysts-bot-data/daily-papers-stats and the Daily Papers API. Open data: tardellirs/paper-pulse-data. Sister project: Model Pulse, the download history of every model on the Hub.