发表机构
RWTH Aachen University; Fraunhofer FIT（亚琛工业大学; 弗劳恩霍夫应用信息技术研究所）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究针对时序异质图学习中跨类型迁移与关系专业化难以兼顾、时间注入方式局限的问题，提出双分支THGFM模型，通过双路径融合与旋转时序注意力，在多学术图基准上取得显著性能提升。
AI 中文摘要
时序异质图为包含多种节点与关系类型且随时间演化的动态关系系统提供了自然的抽象表示。对这类图的学习需要联合建模跨类型结构异质性与交互的时序动态性，但现有方法仍难以兼顾参数高效的跨类型迁移与关系感知的专业化，且通常仅将时间作为注意力核之外的附加特征注入。我们提出THGFM，这是一种网页级时序异质图融合模型，通过统一的双路径架构解决上述两个局限。THGFM将用于参数高效跨类型迁移的共享空间时序注意力分支，与用于关系感知专业化的关系类型分区时序注意力分支耦合，并通过双路径关系-共享融合机制整合，该机制以类型条件非竞争门控和融合实现，这是一种自适应机制，为共享分支与专业化分支分配独立的、类型条件的逐特征门控，允许两者被放大或抑制而无需零和竞争。为将相对时间直接融入注意力得分，THGFM进一步引入旋转时序注意力，其在匹配前按相对时间的半相位旋转查询与键。THGFM在学术图基准上始终优于基线图Transformer模型，实现了+3.25%的六任务平均增益，在OAG-CS PV上的峰值相对增益为+12.37%，在PF-L₂上为+4.87%，在PF-L₁上为+1.18%，在OGBN-MAG、HTAG-ArXiv和HTAG-DBLP上分别为+4.24%、+3.73%和+4.61%。
英文摘要
Temporal heterogeneous graphs offer a natural abstraction for dynamic relational systems in which diverse node and relation types co-exist and evolve over time. Learning on such graphs requires jointly modeling cross-type structural heterogeneity and the temporal dynamics of interactions, yet existing methods still struggle to reconcile parameter-efficient cross-type transfer with relation-aware specialization, and typically inject time only as additive features outside the attention kernel. We propose \textbf{THGFM}, a web-scale temporal heterogeneous graph fusion model that addresses both limitations within a unified dual-path architecture. THGFM couples a \textit{Shared-Space Temporal Attention} branch for parameter-efficient cross-type transfer with a \textit{Relational Type-Partitioned Temporal Attention} branch for relation-aware specialization, and integrates them through \textit{Dual-Path Relational--Shared Fusion}, instantiated with \textit{Type-Conditioned Non-Competitive Gated Sum Fusion}: a adaptive mechanism that assigns independent, type-conditioned feature-wise gates to the shared and specialized branches, allowing both to be amplified or suppressed without zero-sum competition. To directly incorporate relative time into the attention score, THGFM further introduces \textit{Rotary Temporal Attention}, which rotates queries and keys by half-phases of relative time before matching. THGFM consistently outperforms baseline graph transformer models on academic graphs benchmarks, delivering a $+3.25\%$ six-task mean gain, with peak relative gains of $+12.37\%$ on OAG-CS PV, $+4.87\%$ on PF-$L_2$, and $+1.18\%$ on PF-$L_1$, and $+4.24\%$, $+3.73\%$, and $+4.61\%$ on OGBN-MAG, HTAG-ArXiv, and HTAG-DBLP, respectively.
CommentsAccepted at the 25th International Semantic Web Conference (ISWC 2026), Research Track
