发表机构
South China University of Technology; Meta Superintelligence Lab（华南理工大学; Meta超级智能实验室）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
LoGIC通过预算化上下文构建，利用结构、特征和覆盖通道检索标记节点，在保持性能的同时降低内存需求，实现百万节点图上的冻结图上下文学习。
AI 中文摘要
表格基础模型已成为强大的图学习器。诸如G2T-FM和GraphPFN等系统将每个节点编码为特征行，并通过上下文学习（ICL）进行预测，其中标记行作为提示。当前协议使用完整的训练表作为上下文，导致注意力随标记池呈二次方扩展，并引入预处理和内存瓶颈。我们研究了节点级图ICL的上下文构建：对于特定查询，哪些标记节点和辅助未标记节点应构成提示。我们根据两种资源来表述这种分配：用于预测证据的标记上下文预算，以及用于适配器消息传递而不使用标签容量的未标记光环预算。我们提出了LoGIC，它通过结构、基于特征和覆盖通道检索标记节点，在图局部簇中的查询之间共享每个上下文，为适配器主干整合未标记光环，并在没有测试标签的情况下选择通道和上下文预算。在GraphLand上来自两个模型族的三种主干配置中，预算上下文保持了本地可运行的完整上下文性能，与已发布的大数据集结果保持竞争力，并显著降低了与完整上下文和全图推理相比的峰值内存需求。它们还允许在百万节点图上进行冻结图ICL而无需重新训练。我们的分析确定了检索通道何时表现最佳，并将其行为与图属性联系起来。
英文摘要
Tabular foundation models have become powerful graph learners. Systems such as G2T-FM and GraphPFN encode each node as a feature row and make predictions through in-context learning (ICL), with labeled rows serving as the prompt. Current protocols employ the complete training table as context, causing attention to scale quadratically with the labeled pool and introducing preprocessing and memory bottlenecks. We investigate context construction for node-level graph ICL: which labeled nodes and auxiliary unlabeled nodes should constitute the prompt for specified queries. We formulate this allocation in terms of two resources: a labeled-context budget for predictive evidence and an unlabeled-halo budget for adapter message passing without using label capacity. We present LoGIC, which retrieves labeled nodes via structural, feature-based, and coverage channels, shares each context across the queries in a graph-local cluster, incorporates an unlabeled halo for adapter backbones, and chooses the channel and context budget without test labels. Across three backbone configurations drawn from two model families on GraphLand, budgeted contexts maintain locally runnable full-context performance, stay competitive with published large-dataset results, and markedly lower peak memory requirements compared with full-context and whole-graph inference. They further permit frozen graph ICL on million-node graphs without retraining. Our analysis identifies when retrieval channels work best and connects their behavior with graph properties.
