用注意力机制选最佳初始噪声，让视频生成更稳定清晰。
Model Already Knows the Best Noise: Bayesian Active Noise Selection via Attention in Video Diffusion Model
- 通过多组注意力采样计算熵分歧，评估模型对噪声的不确定性。
- 在多个文本到视频模型上提升画质与时间连贯性，推理开销极低。
- 适合追求生成质量稳定的视频扩散模型研究者和开发者。
初始噪声的选择显著影响视频扩散模型的生成质量与提示对齐效果；相同提示下不同随机种子可能产生截然不同的结果。现有方法依赖外部先验（如频率过滤或帧间平滑），却忽略了模型内部信号中蕴含的优选种子线索。为此，我们提出ANSE（生成用主动噪声选择），一种基于模型感知的框架，通过量化注意力不确定性来筛选高质量种子。核心是BANSA（基于注意力的贝叶斯主动噪声选择），其利用多组随机注意力采样的熵分歧衡量模型置信度与一致性。为实现高效推理部署，我们引入伯努利掩码近似版BANSA，仅需单步扩散和部分关键注意力层即可估算得分。在多种文本到视频骨干网络上的实验表明，该方法在几乎无额外推理开销下显著提升了视频质量与时间连贯性，提供了一种原则性强且通用的噪声选择方案。
原文摘要 · Abstract (English)
The choice of initial noise strongly affects quality and prompt alignment in video diffusion; different seeds for the same prompt can yield drastically different results. While recent methods use externally designed priors (e.g., frequency filtering or inter-frame smoothing), they often overlook internal model signals that indicate inherently preferable seeds. To address this, we propose ANSE (Active Noise Selection for Generation), a model-aware framework that selects high-quality seeds by quantifying attention-based uncertainty. At its core is BANSA (Bayesian Active Noise Selection via Attention), an acquisition function that measures entropy disagreement across multiple stochastic attention samples to estimate model confidence and consistency. For efficient inference-time deployment, we introduce a Bernoulli-masked approximation of BANSA that estimates scores from a single diffusion step and a subset of informative attention layers. Experiments across diverse text-to-video backbones demonstrate improved video quality and temporal coherence with marginal inference overhead, providing a principled and generalizable approach to noise selection in video diffusion. See our project page: https://anse-project.github.io/anse-project/.
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。