初始化即半程：从引导势后验生成多样图像
Initialization is Half the Battle: Generating Diverse Images from a Guidance Potential Posterior
AI总结：
针对生成模型模式崩溃问题，提出从引导势后验中采样初始噪声的DivIn方法，利用朗之万动力学引导初始化远离崩溃区域，提升多样性且兼容扩散与流匹配模型。
摘要：
尽管生成模型具有显著的保真度，但它们经常遭受模式崩溃。现有的增强多样性的策略主要集中于在生成轨迹期间进行干预。我们发现一个关键的疏忽：标准高斯初始化通常导致轨迹崩溃到主导模式，因为它对引导势景观是无关的。在这项工作中，我们从引导势后验中公式化选择初始噪声，这有效地将先验重新加权到多样性丰富的区域。为了高效地从该分布中采样，我们引入了多样性诱导初始化（DivIn），它利用朗之万动力学主动导航初始化景观，将初始噪声引导远离崩溃区域，同时将其锚定到有效的数据流形。我们的方法作为一种推理时多样性增强，与扩散和流匹配模型都兼容。大量实验表明，DivIn在类到图像和文本到图像场景中都表现出优越的性能。此外，我们强调，由于DivIn与基于轨迹的方法是正交的，将它们结合起来显著扩展了多样性-质量帕累托前沿，超越了任何单独方法所能达到的。
英文摘要：
Despite the remarkable fidelity of generative models, they frequently suffer from mode collapse. Existing strategies for enhancing diversity predominantly focus on intervening during the generation trajectory. We identify a critical oversight that the standard Gaussian initialization often causes trajectories to collapse into dominant modes because it is agnostic to the guidance potential landscape. In this work, we formulate selecting the initial noise from a guidance potential posterior, which effectively re-weights the prior towards diversity-rich regions. To sample from this distribution efficiently, we introduce Diversity-inducing Initialization (DivIn), which leverages Langevin dynamics to actively navigate the initialization landscape, steering initial noise away from collapsing regions while anchoring them to the valid data manifold. Our method serves as an inference-time diversity enhancement compatible with both diffusion and flow matching models. Extensive experiments show that DivIn exhibits a superior performance in both class-to-image and text-to-image scenarios. Furthermore, we highlight that as DivIn is orthogonal to trajectory-based methods, combining them significantly expands the diversity-quality Pareto frontier beyond what either achieves in isolation.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
