Abstract:Text-attributed graphs (TAGs) are widely used in many real-world domains, and learning on TAGs requires jointly modeling text semantics and graph structure. A standard approach for modeling TAGs is to combine a language model (LM) and a graph neural network (GNN), but joint training is computationally expensive and difficult to scale. Dataset distillation is a promising way to reduce training costs, but existing methods are not well suited to TAGs because they are typically designed for a single modality or still require repeatedly training expensive LM-GNN models on the full dataset during distillation. To address this, we propose TaLK, an effective dataset distillation method for TAGs that couples an LM with a graph-aware neural tangent this http URL design enables efficient dataset distillation, avoiding repeated joint training on the full dataset while reflecting both textual and structural information for effective TAG this http URL on multiple TAG benchmarks show that TaLK consistently outperforms existing baselines and achieves up to 97% of full-dataset performance with only 1% synthetic data.
From: Yeongho Kim [view email]
[v1]
Mon, 22 Jun 2026 07:55:55 UTC (574 KB)
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。