---
格式版本: 2
标题: "Sobolev Regularized Score Difference Estimation in Diffusion Models"
原文链接: "https://arxiv.org/abs/2608.18237"
发布日期: "2026-08-18"
发布时间校准状态: "found"
发布时间需复核: "否"
发布时间来源: "rule:local:strict_original_body"
发布时间证据: "**\\[v1\\]** Tue, 18 Aug 2026 18:23:30 UTC (2,225 KB)"
发布时间校准原因: "规则确认唯一严格发布时间，来源 local:strict_original_body"
发布时间校准置信度: "high"
发布时间候选数量: 18
发布时间严格候选数量: 6
发布时间原页读取状态: "source template page reused from URL open"
发布时间未找到原因: ""
发布时间校准时间: "2026-08-20T14:22:51+08:00"
发布时间仲裁状态: "skipped"
发布时间仲裁尝试次数: 0
发布时间仲裁耗时毫秒: 0
发现时间: "2026-08-20T14:19:37+08:00"
入库时间: "2026-08-20T06:22:52.089Z"
来源平台: "arXiv 学术论文搜索"
搜索渠道: "source_template"
搜索词: "https://arxiv.org/search/?query=Scale-up&searchtype=all"
匹配关键词:
  - "Scale-up"
  - "performance"
相关厂家:
  []
相关专家:
  []
内容类型: "网页"
抓取工具: "Free Fetch + Defuddle"
清洗工具: "Defuddle Markdown + Defuddle/Readability 正文提取"
原始附件:
  []
AI优质: "否"
AI打分: 2
AI分档: "非优质"
AI质检状态: "不通过"
AI打分理由: "论文主题为扩散模型中的分数差估计，属机器学习数学研究，与超节点/AI Rack/机柜级AI基础设施完全无关。"
AI质检模型: "ali-deepseek-v4-flash"
AI质检时间: "2026-08-20T14:23:44+08:00"
AI主题相关性: 0
AI来源权威性: 1
AI新颖性: 0
AI技术细节: 0
AI商业部署信号: 0
AI完整性: 1
AI摘要: "本文提出一种基于Sobolev正则化的分数差估计器，用于扩散模型中的迁移学习与判别器引导，兼具统计一致性和高维可扩展性；实验表明在ECG信号生成等小样本迁移任务上，其稳定性与下游分类性能显著优于非正则化方法。"
AI摘要模型: "ali-deepseek-v4-flash"
AI摘要时间: "2026-09-07T03:21:41.859Z"
采集批次: "2026年8月20日14点19分32秒"
采集批次ID: "20260820-141932-079"
去重键: "https://arxiv.org/abs/2608.18237"
---

## Statistics > Machine Learning

## Title:Sobolev Regularized Score Difference Estimation in Diffusion Models

Authors:[Chenghan Xie](https://arxiv.org/search/stat?searchtype=author&query=Xie,+C), [Jose Blanchet](https://arxiv.org/search/stat?searchtype=author&query=Blanchet,+J), [Renyuan Xu](https://arxiv.org/search/stat?searchtype=author&query=Xu,+R)

[View PDF](https://arxiv.org/pdf/2608.18237) [HTML (experimental)](https://arxiv.org/html/2608.18237v1)

> Abstract:Estimating the difference of two Stein's score functions is a fundamental problem in generative modeling. In particular, score differences arise naturally in transfer learning, where the score difference provides the mechanism for adapting a pre-trained model to a new target distribution, and in diffusion model-based post-training methods such as discriminator guidance. Existing estimators for score differences in these settings either lack of statistical consistency or are difficult to scale up in high-dimensions. We propose a statistically consistent and scalable estimator for score differences based on Sobolev regularization, which plays a crucial role in ensuring consistency and stablizing the training in the small-sample regime. Mathematically, we establish a convergence rate of $O(n^{-\frac{s-1}{d+2s-2}})$ where $d$ is the dimension and $s$ denotes the smoothness of the underlying densities, and provide a minimax lower bound of $\tilde{\Omega}(n^{-\frac{2(s-1)}{d+2s}})$ (in mean-squared error). Empirically, our estimator exhibits significantly improved stability in small-sample regimes compared to existing methods. We demonstrate its effectiveness on real-world tasks, including transfer learning for ECG signal generation, where it substantially outperforms non-regularized score difference estimators in downstream classification performance.

| Comments: |  |
| --- | --- |
| Subjects: | Machine Learning (stat.ML); Machine Learning (cs.LG); Optimization and Control (math.OC) |
| Cite as: | [arXiv:2608.18237](https://arxiv.org/abs/2608.18237) \[stat.ML\] |
|  | (or [arXiv:2608.18237v1](https://arxiv.org/abs/2608.18237v1) \[stat.ML\] for this version) |
|  | [https://doi.org/10.48550/arXiv.2608.18237](https://doi.org/10.48550/arXiv.2608.18237) |

## Submission history

From: Chenghan Xie \[[view email](https://arxiv.org/show-email/44e88853/2608.18237)\]  
**\[v1\]** Tue, 18 Aug 2026 18:23:30 UTC (2,225 KB)

[Which authors of this paper are endorsers?](https://arxiv.org/auth/show-endorsers/2608.18237) | Disable MathJax ([What is MathJax?](https://info.arxiv.org/help/mathjax.html))