Inspired by the great success of foundation models in the development of artificial general intelligence, Graph Foundation Models (GFMs) have recently attracted significant attention and have the potential to advance graph intelligence. Exhibiting such intelligence requires effectively learning knowledge from large-sca...
Yu-Peng Tong, Lianze Shan, Ji-Tao Zhao et al.· 0 citations
Large Language Models (LLMs) have shown strong potential for text-attributed graph learning by leveraging rich pre-trained knowledge and language understanding capabilities. Combining LLMs with Graph Neural Networks (GNNs) offers a promising paradigm for expanding the modeling capabilities of graph learning. However, L...
Pei-Yuan Li, Yong-Qi Huang, Ji-Tao Zhao et al.· 0 citations
SP decomposes graph information into Chebyshev polynomial bases and learns shared spectral responses, enabling unified yet adaptive cognition of continuously varying spectral patterns across graphs and enables effective and transferable graph cognition.
Ankang Yang, Ji-Tao Zhao, Dong-Xiao He et al.· 0 citations
A multi-semantic basis representation learning paradigm that models each multi-label node as an adaptive composition of semantic bases, thereby enabling flexible representational capacity for modeling multiple semantics.
Dong-Xiao He, Jia-Yu Zhang, Ji-Tao Zhao et al.· 0 citations
CHARM is proposed, a multimodal graph foundation model with hierarchical context modeling for zero-shot transfer that replaces isolated raw nodes with hierarchical graph contexts that capture multimodal semantics and cross-modal relations.
Ankang Yang, Ji-Tao Zhao, Di Jin et al.· arXiv.org· 0 citations
SliGFM is proposed, a graph foundation model built upon topology-aware sliding-window feature encoding and generative reconstruction that enables a smoothness-aware transformer to capture transferable relational patterns among feature tokens within each node, while the generative reconstruction objective encourages pre...
A Propagation-aware Graph Foundation Model (ProGFM), which regards the propagation relationships between edges and feature dimensions as transferable knowledge units, and exhibits superior generalization performance compared with existing methods.
Yi Wang, Jitao Zhao, Di Jin et al.· arXiv.org· 0 citations
HGMRP extends single-relation prompts into a composite structure that includes both relation-specific and shared components, thereby enhancing expressive capability and significantly out-performs existing heterogeneous prompt methods, validating its effectiveness and superiority.
Feng-Yu Yan, Xiao-Bao Wang, Qian-Xi Tang et al.· Proceedings of the Thirty-Fi...· 0 citations
This work proposes AgentGFM, in which all node agents follow a shared end-to-end trainable policy rather than using independent models, and describes this capability as information-flow control, which is inspired by recent advances in agent technology.
Jingbo Cui, Jitao Zhao, Di Jin et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
