通过测试时进化搜索实现图像与视频生成的规模化扩展

Scaling Image and Video Generation via Test-Time Evolutionary Search

May 23, 2025

作者: Haoran He, Jiajun Liang, Xintao Wang, Pengfei Wan, Di Zhang, Kun Gai, Ling Pan

cs.AI

摘要

随着模型预训练阶段扩展计算资源（数据和参数）的边际成本持续显著上升，测试时扩展（TTS）作为一种在推理阶段分配额外计算以提升生成模型性能的途径，展现出广阔前景。尽管TTS在多项语言任务中已取得显著成效，但对于图像和视频生成模型（基于扩散或流模型）的测试时扩展行为理解仍存在明显空白。虽然近期研究已开始探索视觉任务的推理时策略，但这些方法面临关键局限：局限于特定任务领域、可扩展性差，或陷入奖励过度优化而牺牲样本多样性。本文提出进化搜索（EvoSearch），一种新颖、通用且高效的TTS方法，无需额外训练或模型扩展，即可有效增强扩散和流模型在图像与视频生成上的可扩展性。EvoSearch将扩散和流模型的测试时扩展重构为进化搜索问题，借鉴生物进化原理，高效探索并优化去噪轨迹。通过精心设计针对随机微分方程去噪过程的选择与变异机制，EvoSearch在保持种群多样性的同时，迭代生成更高质量的后代。通过对图像和视频生成任务中扩散与流架构的广泛评估，我们证明该方法持续超越现有方法，实现更高多样性，并对未见评估指标展现出强大的泛化能力。项目详情请访问https://tinnerhrhe.github.io/evosearch。

English

As the marginal cost of scaling computation (data and parameters) during model pre-training continues to increase substantially, test-time scaling (TTS) has emerged as a promising direction for improving generative model performance by allocating additional computation at inference time. While TTS has demonstrated significant success across multiple language tasks, there remains a notable gap in understanding the test-time scaling behaviors of image and video generative models (diffusion-based or flow-based models). Although recent works have initiated exploration into inference-time strategies for vision tasks, these approaches face critical limitations: being constrained to task-specific domains, exhibiting poor scalability, or falling into reward over-optimization that sacrifices sample diversity. In this paper, we propose Evolutionary Search (EvoSearch), a novel, generalist, and efficient TTS method that effectively enhances the scalability of both image and video generation across diffusion and flow models, without requiring additional training or model expansion. EvoSearch reformulates test-time scaling for diffusion and flow models as an evolutionary search problem, leveraging principles from biological evolution to efficiently explore and refine the denoising trajectory. By incorporating carefully designed selection and mutation mechanisms tailored to the stochastic differential equation denoising process, EvoSearch iteratively generates higher-quality offspring while preserving population diversity. Through extensive evaluation across both diffusion and flow architectures for image and video generation tasks, we demonstrate that our method consistently outperforms existing approaches, achieves higher diversity, and shows strong generalizability to unseen evaluation metrics. Our project is available at the website https://tinnerhrhe.github.io/evosearch.