提出新型生成模型，一步到多步都能高效生成高质量图像。
Transition Models: Rethinking the Generative Learning Objective
- 用连续时间方程建模任意时间间隔的状态转移，突破传统生成目标限制。
- 865M参数模型在所有采样步数下超越8B/12B参数的顶尖模型，且质量随步数提升。
- 支持最高4096x4096分辨率生成，适合需要高质高效图像生成的场景。
生成建模中存在根本性矛盾：迭代扩散模型虽保真度高但计算成本大，而高效少步方案则受限于质量天花板。根源在于训练目标仅关注极小时间动态（PF-ODE）或直接端点预测。本文提出一个精确的连续时间动力学方程，可解析定义任意有限时间间隔内的状态转移。由此构建的新一代生成范式——过渡模型（TiM），能自适应任意步数的生成过程，无缝覆盖从单步跳跃到精细优化的整个轨迹。尽管仅有865M参数，TiM在所有评估步数下均超越主流模型如SD3.5（8B参数）和FLUX.1（12B参数）。更重要的是，不同于以往少步生成器，TiM在增加采样预算时表现出单调的质量提升。此外，采用原生分辨率策略后，其在4096x4096分辨率下仍保持卓越保真度。
原文摘要 · Abstract (English)
A fundamental dilemma in generative modeling persists: iterative diffusion models achieve outstanding fidelity, but at a significant computational cost, while efficient few-step alternatives are constrained by a hard quality ceiling. This conflict between generation steps and output quality arises from restrictive training objectives that focus exclusively on either infinitesimal dynamics (PF-ODEs) or direct endpoint prediction. We address this challenge by introducing an exact, continuous-time dynamics equation that analytically defines state transitions across any finite time interval. This leads to a novel generative paradigm, Transition Models (TiM), which adapt to arbitrary-step transitions, seamlessly traversing the generative trajectory from single leaps to fine-grained refinement with more steps. Despite having only 865M parameters, TiM achieves state-of-the-art performance, surpassing leading models such as SD3.5 (8B parameters) and FLUX.1 (12B parameters) across all evaluated step counts. Importantly, unlike previous few-step generators, TiM demonstrates monotonic quality improvement as the sampling budget increases. Additionally, when employing our native-resolution strategy, TiM delivers exceptional fidelity at resolutions up to 4096x4096.
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。