The paper introduces CoTeach, a confidence‑aware dual‑teacher learning framework for few‑shot node classification on text‑attributed graphs. It dynamically chooses between a Graph Neural Network and a Large Language Model as the teacher for each node, based on which source is more reliable for that node. Experiments show that this approach improves classification accuracy while cutting unnecessary use of expensive LLMs.
By Hojin Kim, Sujin Yoon, Sungsu Lim, Dongwon Lee, David Yoon Suk Kang
The paper examines a multimodal approach that combines a self‑supervised GNN encoder with an alternating optimization scheme involving a language‑model teacher. Despite the expectation that this joint strategy would enhance predictive performance, the authors find that the combined model fails to deliver significant gains. They identify six key factors—ranging from anchor strength trade‑offs to misaligned representation spaces—that explain why the integration of text knowledge does not fully benefit graph learning.
By Fumiaki Kimino (SOKENDAI), Ryoma Sato (SOKENDAI, National Institute of Informatics)
PACE introduces a propagation‑aware collaborative correction for one‑shot personalized federated graph learning. Each client sends a rank‑r update and a diagonal sketch of message moments, allowing the server to construct a correction that anchors to the receiver’s local model. A convex negative‑log‑likelihood calibration selects a single coefficient to blend local and external logits, improving accuracy and weighted‑F1 on most datasets while preserving local predictions when the correction is unhelpful.
By Ruizhe Huang, Chengran Li, Xiaochuan Shi
arXiv:2606. 10249v1 Announce Type: new Abstract: We examine whether graph neural network (GNN) design rules generalize across benchmark families by studying aggregator selection (sum, mean, max) on 24 node-classification datasets spanning citation, heterophilic, LINKX Facebook-100, co-purchase, and co-authorship graphs.
By Neha Sharma, Ritesh Sharma
arXiv:2609.37057v1 Announce Type: new
Abstract: Achieving strong performance with graph neural networks (GNNs) typically requires training and hyperparameter tuning for each dataset, incurring repeat...
By Dooho Lee, Jinmo Lee, Minho Jeong, Kijung Shin, Jaemin Yoo
Text-attributed graphs (TAGs) are an important graph data form that combine relational structure with rich node text. However, real-world TAGs are often imperfect, with quality issues arising from text, structure, and labels, and typically manifesting as sparsity, noise, and imbalance.
arXiv:2607. 13643v1 Announce Type: cross Abstract: Sampling multiple solutions and returning the majority answer is among the most reliable ways to improve the reasoning accuracy of large language models without labels, and a growing family of methods converts this consensus signal into training supervision.
By John Gkountouras, Josip Juki\'c, Ivan Titov
The paper introduces GONE, a benchmark for evaluating knowledge unlearning in large language models using structured knowledge graphs, and presents Neighborhood-Expanded Distribution Shaping (NEDS), a framework that leverages graph connectivity to separate forgotten facts from their semantic neighborhood. GONE disentangles direct fact removal, reasoning-based leakage, and catastrophic forgetting, while NEDS achieves high unlearning efficacy and locality on LLaMA-3-8B and Mistral-7B. The dataset is publicly available on Hugging Face.
By Chahana Dahal, Ashutosh Balasubramaniam, Zuobin Xiong
arXiv:2607. 21885v1 Announce Type: new Abstract: Coarsening-based training for graph neural networks (GNNs), i.
By Guoming Li, Jian Yang, Xukun Wang, Zixiao Wang, Shangsong Liang, Yifan Chen
arXiv:2606. 12863v2 Announce Type: replace Abstract: Multimodal attributed graphs (MAGs) integrate graph topology with heterogeneous modality attributes, such as text and images, thereby enabling richer modeling of complex relational systems.
By Zhengyu Wu, Xu Wang, Hongchao Qin, Xunkai Li, Guang Zeng, Rong-Hua Li, Guoren Wang
arXiv:2608. 11431v1 Announce Type: new Abstract: Graph learning presupposes a graph, and tables and relational databases do not come with one.
By Tamara Cucumides, Floris Geerts
Sampling multiple solutions and returning the majority answer is among the most reliable ways to improve the reasoning accuracy of large language models without labels, and a growing family of methods converts this consensus signal into training supervision. However, existing approaches use consensus only in restricted forms: as a filter that selects solutions for fine-tuning, as a preference between answers, or as a scalar reward for reinforcement learning, discarding most of the information that the agreeing solutions contain.