Learning a Prior-Data Fitted Network for Graph Foundation Models via Structural Causal Model Priors

这个问题是要把一段纯文本新闻正文排版成语义化的HTML片段，得遵守用户给的规则。首先看正文内容，是关于Graph Foundation Models的研究，里面有几个长段落，还有加粗的重要术语GSCM-PFN，以及代码链接。首先，每个自然段落要用

包裹，这里正文有三个自然段落吗？不对，看原文：第一段讲GFMs的现状和局限，第二段讲PFNs的启发和现有graph PFNs的问题，第三段是提出的GSCM-PFN及其三个贡献，最后还有实验结果和代码链接。不过用户给的正文里，第三段比较长，包含具体的方法和实验结果。然后，加粗的内容要保留，原文里GSCM-PFN是GSCM-PFN，所以HTML里要用GSCM-PFN。还有代码链接，原文里是Code: https://...，按照规则，代码用

，所以要把链接放里面吗？看规则里代码用
，所以应该是
https://anonymous.4open.science/r/GSCM-PFN-A3A3/
。接下来检查每个自然段：第一个自然段是“Graph Foundation Models...textual information in Large Language Models (LLMs).”用
包裹；第二个自然段是“To overcome these limitations...complex synthetic datasets.”用
；第三个自然段是“To address these challenges...prior that jointly models nodes and graph topology.”用
；然后接下来的三个贡献点，原文是“Specifically, we (1) design...(2) introduce...and (3) propose...”，这部分是列表内容，所以要用
，每个(1)(2)(3)作为
的内容；然后最后一句“Experiments on 23 real-world benchmarks...Code: https://...”里，前面的实验结果部分是自然段吗？不，看原文，“Experiments on 23 real-world benchmarks show that GSCM-PFN outperforms both conventional GFMs and PFN-based GFMs, without utilizing real-world graph data for pre-training. Code: https://anonymous.4open.science/r/GSCM-PFN-A3A3/”这部分，前面的实验结果是一句话，后面是代码链接，所以要把前面的实验结果用
吗？或者看原文结构：整个正文除了三个贡献点，其余是自然段落。再仔细看用户给的正文：

第一自然段：Graph Foundation Models (GFMs)...textual information in Large Language Models (LLMs).

第二自然段：To overcome these limitations...complex synthetic datasets.

第三自然段：To address these challenges...jointly models nodes and graph topology.

然后是Specifically, we (1)...(2)...and (3)...这部分是列表，所以用
；

接下来是：Experiments on 23 real-world benchmarks...Code: https://...这部分，前面的实验结果是一句话，后面是代码链接，所以要把实验结果那句用
，代码链接用
。

现在开始排版：

首先第一自然段：
Graph Foundation Models (GFMs) have recently emerged as a promising paradigm for learning transferable representations across different graphs and domains. However, existing GFMs face notable limitations, including limited pre-training scale or reliance on aligning with textual information in Large Language Models (LLMs).

第二自然段：
To overcome these limitations, recent studies have drawn inspiration from prior-data fitted networks (PFNs), which learn from large synthetic datasets and support multi-domain prediction backed by causal learning. However, existing graph PFNs encounter severe difficulties in modeling causal relationships within and across node and edge sets, generating diverse graph structures, and designing suitable architectures that can learn from complex synthetic datasets.

第三自然段：
To address these challenges, we propose GSCM-PFN (Graph SCM-based Prior-Data Fitted Network), a graph PFN pre-trained on synthetic data generated from an SCM-based prior that jointly models nodes and graph topology.

然后是列表部分：
design a joint feature-edge-label structural causal model (SCM) to guide synthetic graph generation and capture causal relationships within and across node and edge sets;
introduce prior-guided node-edge dependency modeling to instantiate diverse graph generation strategies, and prove that our SCM-based generation prior is more expressive than existing priors; and
propose a feature-permutation-invariant dual-path graph network with theoretical guarantees for learning from diverse synthetic graphs.

接下来是实验结果和代码：
Experiments on 23 real-world benchmarks show that GSCM-PFN outperforms both conventional GFMs and PFN-based GFMs, without utilizing real-world graph data for pre-training. Code:
https://anonymous.4open.science/r/GSCM-PFN-A3A3/

检查是否保留所有信息，有没有增删改：原文里的三个贡献点的(1)(2)(3)在列表里有没有去掉？原文是“we (1) design...(2) introduce...and (3) propose...”，所以列表里的
要包含这些序号吗？看规则里列表内容用
，所以应该保留，比如
(1) design a joint feature-edge-label structural causal model (SCM) to guide synthetic graph generation and capture causal relationships within and across node and edge sets;
，这样才完整。哦对，原文里的(1)(2)(3)不能丢，所以刚才的列表
里要加上这些序号。

所以调整列表部分：

(1) design a joint feature-edge-label structural causal model (SCM) to guide synthetic graph generation and capture causal relationships within and across node and edge sets;
(2) introduce prior-guided node-edge dependency modeling to instantiate diverse graph generation strategies, and prove that our SCM-based generation prior is more expressive than existing priors; and
(3) propose a feature-permutation-invariant dual-path graph network with theoretical guarantees for learning from diverse synthetic graphs.

然后实验结果那句里的Code:后面接代码块，所以是
Experiments on 23 real-world benchmarks show that GSCM-PFN outperforms both conventional GFMs and PFN-based GFMs, without utilizing real-world graph data for pre-training. Code:
https://anonymous.4open.science/r/GSCM-PFN-A3A3/

再检查加粗部分，原文里的GSCM-PFN要变成GSCM-PFN，没错。

有没有长文用
/
划分小节？原文里的长文有没有需要划分小节的？比如第三段后面的三个贡献点，是不是属于一个小节？原文里“To address these challenges...topology.”之后是“Specifically, we (1)...(3)...”，这部分可以作为
吗？看规则2，长文用
/
划分小节，那这里的内容属于“GSCM-PFN的具体设计”小节，所以可以用
吗？原文里没有小节标题，所以不能自己加标题，规则4说不得增删改内容，所以不能添加
/
，因为没有原文的小节标题，所以只能用
和

               。 所以最终的HTML片段应该是： 
               
Graph Foundation Models (GFMs) have recently emerged as a promising paradigm for learning transferable representations across different graphs and domains. However, existing GFMs face notable limitations, including limited pre-training scale or reliance on aligning with textual information in Large Language Models (LLMs).
               
To overcome these limitations, recent studies have drawn inspiration from prior-data fitted networks (PFNs), which learn from large synthetic datasets and support multi-domain prediction backed by causal learning. However, existing graph PFNs encounter severe difficulties in modeling causal relationships within and across node and edge sets, generating diverse graph structures, and designing suitable architectures that can learn from complex synthetic datasets.
               
To address these challenges, we propose GSCM-PFN (Graph SCM-based Prior-Data Fitted Network), a graph PFN pre-trained on synthetic data generated from an SCM-based prior that jointly models nodes and graph topology.
               

                
(1) design a