/
划分小节”，但如果原文没有标题，那怎么划分？可能原文是几个自然段落，比如第一个大段落讲问题障碍，第二个讲现有方法失败，第三个讲ablation研究，第四个讲Cheesefill，第五个讲理论，第六个讲实验。但原文是连续的吗？再看原文： 原文是连续的一段吗？不，看标点，其实是一段，但可能需要分成几个p？不对，规则说每个自然段落用
包裹，原文的自然段落就是一段吗？不对，再仔细读原文，发现原文是： 1. 第一句到第三个障碍讲完，然后是“Existing approaches fail to address these challenges concurrently...” 是另一个部分，然后“Crucially, we show through ablation studies...” ，然后“To overcome this, we introduce Cheesefill...” ，然后“Theoretical...” ，然后“Extensive experiments...” 。所以这些是不同的自然段落吗？原文的自然段落可能是这些部分，所以需要用p包裹每个自然段落。但原文的呈现是一个连续的文本，没有换行，所以需要判断自然段落的划分。 哦，原文的纯文本是： Discovering causal structures in multivariate time series (MTS) is critical in domains such as finance and neuroscience, yet remains challenging in practice due to three distinct obstacles. First, MTS are routinely corrupted by missing values, where naive impute-then-discover pipelines distort the underlying data distribution and yield spurious causal graphs. Second, identifying instantaneous effects is difficult even on complete data, and missingness further entangles imputation with graph recovery. Third, real-world data frequently exhibits heteroscedastic noise with variance driven by both instantaneous and lagged causes, violating the core assumptions of standard discovery methods. Existing approaches fail to address these challenges concurrently, typically presuming complete data, homoscedastic noise, or strictly time-lagged relationships. Crucially, we show through ablation studies that resolving these issues sequentially, rather than jointly, leads to a substantial degradation in performance. To overcome this, we introduce Cheesefill, an optimal transport framework for causal structure learning under missingness and heteroscedastic noise. Cheesefill parameterizes causal mechanisms using conditional normalizing flows and jointly learns a stochastic correction map that refines naive imputations into trajectories consistent with the inferred causal mechanisms. Theoretically, we prove that optimizing over stochastic correction maps is equivalent to a conditional kernel reformulation of the Kantorovich problem, from which our neural parameterization yields a tractable objective. Extensive experiments on synthetic and real-world benchmarks demonstrate that Cheesefill consistently outperforms existing baselines across diverse missingness regimes. 所以原文是一个连续的段落，但里面有几个完整的句子群，可能需要分成几个p？比如： - 第一个p包含前三个障碍的描述：“Discovering... standard discovery methods.” - 第二个p包含现有方法的不足：“Existing approaches fail...” - 第三个p包含ablation研究：“Crucially, we show...” - 第四个p包含Cheesefill框架的介绍：“To overcome this...” - 第五个p包含理论证明：“Theoretical...” - 第六个p包含实验结果：“Extensive experiments...” 但需要确认是否是自然段落。自然段落通常是独立的意义单元，所以这些可以作为自然段落。然后看是否有需要用h2/h3的地方，但原文没有标题，所以可能不需要，除非根据内容划分小节。不过规则允许长文用h2/h3划分小节，但如果原文没有标题，添加的话会增删内容吗？规则说不得增删改内容，所以不能添加标题，只能按原文的自然段落用p包裹。 然后看是否有列表、引用、代码，这里没有，所以不需要。重要句子可以用strong，比如可能有三个障碍的关键点，但原文中是否有重要句子？比如“three distinct obstacles”后面的内容，或者“Cheesefill”是重要的，但需要看是否需要强调。比如“Cheesefill”是框架名称，可能需要用strong吗？规则说重要句子可用strong，所以可以。 现在开始排版： 第一个自然段落：“Discovering causal structures in multivariate time series (MTS) is critical in domains such as finance and neuroscience, yet remains challenging in practice due to three distinct obstacles. First, MTS are routinely corrupted by missing values, where naive impute-then-discover pipelines distort the underlying data distribution and yield spurious causal graphs. Second, identifying instantaneous effects is difficult even on complete data, and missingness further entangles imputation with graph recovery. Third, real-world data frequently exhibits heteroscedastic noise with variance driven by both instantaneous and lagged causes, violating the core assumptions of standard discovery methods.” 包裹在
里。 第二个自然段落：“Existing approaches fail to address these challenges concurrently, typically presuming complete data, homoscedastic noise, or strictly time-lagged relationships.” 包裹在
里。 第三个自然段落：“Crucially, we show through ablation studies that resolving these issues sequentially, rather than jointly, leads to a substantial degradation in performance.” 包裹在
里。 第四个自然段落：“To overcome this, we introduce Cheesefill, an optimal transport framework for causal structure learning under missingness and heteroscedastic noise. Cheesefill parameterizes causal mechanisms using conditional normalizing flows and jointly learns a stochastic correction map that refines naive imputations into trajectories consistent with the inferred causal mechanisms.” 这里Cheesefill是重要名称，用strong。 第五个自然段落：“Theoretical, we prove that optimizing over stochastic correction maps is equivalent to a conditional kernel reformulation of the Kantorovich problem, from which our neural parameterization yields a tractable objective.” 包裹在
里。 第六个自然段落：“Extensive experiments on synthetic and real-world benchmarks demonstrate that Cheesefill consistently outperforms existing baselines across diverse missingness regimes.” 这里再次强调Cheesefill。 但等一下，原文中的“Theoretical”后面有没有逗号？原文是“Theoretically”，打错了，需要保留原文，所以是“Theoretically”。 现在检查是否保留所有信息，没有增删改。然后每个自然段落用
，重要名称用strong。 所以最终的HTML片段应该是：
Discovering causal structures in multivariate time series (MTS) is critical in domains such as finance and neuroscience, yet remains challenging in practice due to three distinct obstacles. First, MTS are routinely corrupted by missing values, where naive impute-then-discover pipelines distort the underlying data distribution and yield spurious causal graphs. Second, identifying instantaneous effects is difficult even on complete data, and missingness further entangles imputation with graph recovery. Third, real-world data frequently exhibits heteroscedastic noise with variance driven by both instantaneous and lagged causes, violating the core assumptions of standard discovery methods.
Existing approaches fail to address these challenges concurrently, typically presuming complete data, homoscedastic noise, or strictly time-lagged relationships.
Crucially, we show through ablation studies that