让扩散模型学会自主规划生成路径，提升质量并省计算。
Chain-of-Trajectories: Unlocking the Intrinsic Generative Optimality of Diffusion Models via Graph-Theoretic Planning
- 用低维签名代替高维噪声空间，将采样转为图规划问题。
- 动态分配计算资源到最难的生成阶段，提升质量和稳定性。
- 无需训练，适合追求高效高质量生成的开发者使用。
扩散模型以固定、无内容感知的采样节奏运行，受限于状态维度灾难，难以进行显式轨迹规划，导致计算资源浪费。为此，我们提出无需训练的链式轨迹（Chain-of-Trajectories, CoTj）框架，实现系统2式的理性规划。核心是扩散DNA——一种量化各阶段去噪难度的低维签名，可代理高维状态空间，使采样转化为有向无环图上的图规划。通过预测-规划-执行范式，CoTj 动态分配计算资源至最困难的生成阶段。在多个生成模型上验证表明，CoTj 能发现上下文感知的生成路径，提升输出质量与稳定性，同时减少冗余计算。本工作为资源感知、基于规划的扩散建模奠定新基础。代码已开源：https://github.com/UnicomAI/CoTj。
原文摘要 · Abstract (English)
Diffusion models operate in a reflexive System 1 mode, constrained by a fixed, content-agnostic sampling schedule. This rigidity arises from the curse of state dimensionality, where the combinatorial explosion of possible states in the high-dimensional noise manifold renders explicit trajectory planning intractable and leads to systematic computational misallocation. To address this, we introduce Chain-of-Trajectories (CoTj), a train-free framework enabling System 2 deliberative planning. Central to CoTj is Diffusion DNA, a low-dimensional signature that quantifies per-stage denoising difficulty and serves as a proxy for the high-dimensional state space, allowing us to reformulate sampling as graph planning on a directed acyclic graph. Through a Predict-Plan-Execute paradigm, CoTj dynamically allocates computational effort to the most challenging generative phases. Experiments across multiple generative models demonstrate that CoTj discovers context-aware trajectories, improving output quality and stability while reducing redundant computation. This work establishes a new foundation for resource-aware, planning-based diffusion modeling. The code is available at https://github.com/UnicomAI/CoTj.
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。