用控制理论加速生成模型采样，减少迭代次数。
PID-controlled Langevin Dynamics for Faster Sampling of Generative Models
- 将能量梯度视为反馈信号，融合历史梯度与趋势变化。
- 图像生成任务中用更少步骤达到更高质量样本。
- 无需训练，可直接集成到现有生成模型中。
Langevin动力学采样因需大量细粒度迭代而生成速度极慢，根本受限于收敛至目标分布所需步数。本文提出PID控制的Langevin动力学（PIDLD），基于控制理论重新诠释采样过程。将能量梯度视为反馈信号，结合历史梯度（积分项）与梯度变化趋势（微分项），高效穿越能量景观并自适应稳定，显著减少生成高质量样本所需的迭代次数。该方法无需额外训练、数据集或先验信息，可立即嵌入任意基于Langevin的方法中。在图像生成与推理任务上的大量实验表明，PIDLD以更少步数实现更高生成质量，使基于Langevin的生成模型在效率敏感场景中更具实用性。代码已公开于 https://github.com/tsinghua-fib-lab/PIDLD。
原文摘要 · Abstract (English)
Langevin dynamics sampling suffers from extremely low generation speed, fundamentally limited by numerous fine-grained iterations to converge to the target distribution. We introduce PID-controlled Langevin Dynamics (PIDLD), a novel sampling acceleration algorithm that reinterprets the sampling process using control-theoretic principles. By treating energy gradients as feedback signals, PIDLD combines historical gradients (the integral term) and gradient trends (the derivative term) to efficiently traverse energy landscapes and adaptively stabilize, thereby significantly reducing the number of iterations required to produce high-quality samples. Our approach requires no additional training, datasets, or prior information, making it immediately integrable with any Langevin-based method. Extensive experiments across image generation and reasoning tasks demonstrate that PIDLD achieves higher quality with fewer steps, making Langevin-based generative models more practical for efficiency-critical applications. The implementation can be found at \href{https://github.com/tsinghua-fib-lab/PIDLD}{https://github.com/tsinghua-fib-lab/PIDLD}.
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。