# Continuous Adversarial MeanFlow Transfer

arXiv: https://arxiv.org/abs/2608.19540 · HTML: https://mlsanity.com/paper/2608.19540/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 25 of its 45 references that are in the map.
- First posted on arXiv 20 Aug 2026.

## Builds on (most cited first, 25 of 25)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [Classifier-Free Diffusion Guidance](https://mlsanity.com/paper/2207.12598/) (2022)
- [Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow](https://mlsanity.com/paper/2209.03003/) (2022)
- [DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generation](https://mlsanity.com/paper/2208.12242/) (2022)
- [One-step Diffusion with Distribution Matching Distillation](https://mlsanity.com/paper/2311.18828/) (2023)
- [Improved Distribution Matching Distillation for Fast Image Synthesis](https://mlsanity.com/paper/2405.14867/) (2024)
- [Mean Flows for One-step Generative Modeling](https://mlsanity.com/paper/2505.13447/) (2025)
- [Consistency Trajectory Models: Learning Probability Flow ODE Trajectory of Diffusion](https://mlsanity.com/paper/2310.02279/) (2023)
- [SVDiff: Compact Parameter Space for Diffusion Fine-Tuning](https://mlsanity.com/paper/2303.11305/) (2023)
- [Fast High-Resolution Image Synthesis with Latent Adversarial Diffusion Distillation](https://mlsanity.com/paper/2403.12015/) (2024)
- [LCM-LoRA: A Universal Stable-Diffusion Acceleration Module](https://mlsanity.com/paper/2311.05556/) (2023)
- [Diffusion Adversarial Post-Training for One-Step Video Generation](https://mlsanity.com/paper/2501.08316/) (2025)
- [Flow map matching with stochastic interpolants: A mathematical framework for consistency models](https://mlsanity.com/paper/2406.07507/) (2024)
- [Align Your Flow: Scaling Continuous-Time Flow Map Distillation](https://mlsanity.com/paper/2506.14603/) (2025)
- [Flash Diffusion: Accelerating Any Conditional Diffusion Model for Few Steps Image Generation](https://mlsanity.com/paper/2406.02347/) (2024)
- [One-step Latent-free Image Generation with Pixel Mean Flows](https://mlsanity.com/paper/2601.22158/) (2026)
- [Diffusion Models without Classifier-free Guidance](https://mlsanity.com/paper/2502.12154/) (2025)
- [Diff2Flow: Training Flow Matching Models via Diffusion Model Alignment](https://mlsanity.com/paper/2506.02221/) (2025)
- [Decoupled MeanFlow: Turning Flow Models into Flow Maps for Accelerated Sampling](https://mlsanity.com/paper/2510.24474/) (2025)
- [Fast Text-to-Audio Generation with Adversarial Post-Training](https://mlsanity.com/paper/2505.08175/) (2025)
- [Diffusion Tuning: Transferring Diffusion Models via Chain of Forgetting](https://mlsanity.com/paper/2406.00773/) (2024)
- [Expanding Expressiveness of Diffusion Models with Limited Data via Self-Distillation based Fine-Tuning](https://mlsanity.com/paper/2311.01018/) (2023)
- [Domain Guidance: A Simple Transfer Approach for a Pre-trained Diffusion Model](https://mlsanity.com/paper/2504.01521/) (2025)
- [Few-Shot Image Generation by Conditional Relaxing Diffusion Inversion](https://mlsanity.com/paper/2407.07249/) (2024)
- [DogFit: Domain-guided Fine-tuning for Efficient Transfer Learning of Diffusion Models](https://mlsanity.com/paper/2508.05685/) (2025)

## Abstract (arXiv)

Training fast generators on new domains with limited data remains challenging for two reasons. First, adapting a pretrained diffusion or flow model to a new domain leaves its costly multi-step sampling unaddressed, and existing acceleration methods are tied to the source parameterization--ε, x, v, or u--leaving heterogeneous pretrained models with no common acceleration target. Second, while adversarial refinement is proven effective for few-step quality, it is formulated only for instantaneous-velocity flows, not for the finite-interval average velocities that MeanFlow (MF) models predict. We address both problems. We propose MeanFlow-Transfer, which maps heterogeneous source outputs into a shared velocity representation, uses it to initialize an MF generator from the source weights, and optimizes an MF objective on the target domain. This unifies adaptation and acceleration in a single training loop across a broad range of pretrained models. We then introduce Continuous Adversarial MeanFlow, a post-training stage that extends continuous adversarial flow models from instantaneous velocities to MF's finite-interval average velocities. CAMF contrasts changes in a learned potential between real and predicted interval endpoints, recovering fine detail that MF regression averages away, and reduces to the instantaneous criterion in the vanishing-interval limit. Adapting four ImageNet-based source models--DiT (ε), SiT (v), JiT (x), iMF (u)--to five target domains, MF-T with CAMF matches or exceeds the fine-tuned teacher in FID and FDD at up to 125× fewer Neural Function Evaluations (NFEs), while CAMF improves MF-T's few-step FID by 29% on average.