arXiv:2505.23343cs.CV2025-05

通过分析采样路径,提前筛选低质图像,无需重训练或外部奖励。

Diffusion Sampling Path Tells More: An Efficient Plug-and-Play Strategy for Sample Filtering

  • 利用条件与无条件得分累积差值识别高质量样本分布区域。
  • 在去噪早期阶段完成过滤,显著提升生成质量,人类偏好评分更高。
  • 即插即用,不改模型结构,适用于各类生成任务。

扩散模型因采样轨迹的随机性常出现样本质量不一致问题。尽管基于训练的微调(如DDPO)和推理时对齐技术旨在提升样本保真度,但通常需完整去噪过程及外部奖励信号,计算开销大。本文揭示一个此前未被关注的发现:在无分类器引导(CFG)下,样本质量与去噪轨迹特征存在强关联。具体而言,样本分布的高密度区与累积得分差异(ASD)密切相关。基于此,我们提出CFG-Rejection策略,可在去噪早期高效过滤低质样本,且无需外部奖励信号或模型重训。该方法无需修改模型架构或采样调度,完全兼容现有扩散框架。在图像生成任务中广泛验证,显著提升人类偏好评分(HPSv2、PickScore)及挑战性基准(GenEval、DPG-Bench)表现。预计该方法可推广至多模态生成任务,推动更高效可靠的高质量样本生成。

原文摘要 · Abstract (English)

Diffusion models often exhibit inconsistent sample quality due to stochastic variations inherent in their sampling trajectories. Although training-based fine-tuning (e.g. DDPO [1]) and inference-time alignment techniques[2] aim to improve sample fidelity, they typically necessitate full denoising processes and external reward signals. This incurs substantial computational costs, hindering their broader applicability. In this work, we unveil an intriguing phenomenon: a previously unobserved yet exploitable link between sample quality and characteristics of the denoising trajectory during classifier-free guidance (CFG). Specifically, we identify a strong correlation between high-density regions of the sample distribution and the Accumulated Score Differences (ASD)--the cumulative divergence between conditional and unconditional scores. Leveraging this insight, we introduce CFG-Rejection, an efficient, plug-and-play strategy that filters low-quality samples at an early stage of the denoising process, crucially without requiring external reward signals or model retraining. Importantly, our approach necessitates no modifications to model architectures or sampling schedules and maintains full compatibility with existing diffusion frameworks. We validate the effectiveness of CFG-Rejection in image generation through extensive experiments, demonstrating marked improvements on human preference scores (HPSv2, PickScore) and challenging benchmarks (GenEval, DPG-Bench). We anticipate that CFG-Rejection will offer significant advantages for diverse generative modalities beyond images, paving the way for more efficient and reliable high-quality sample generation.

扩散模型采样优化即插即用质量筛选

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。