Acacia: New Graph Foundation Model Trained Solely on Web Graph

Researchers have introduced Acacia, a novel graph foundation model trained exclusively on the Common Crawl web graph. Unlike existing models that often require additional training for new tasks or features, Acacia supports arbitrary feature dimensionalities and semantics without retraining. It demonstrates capabilities in node classification, link prediction, node clustering, and graph generation, and notably, it achieves these through in-context learning without relying on pretrained Large Language Models. AI

IMPACT This research demonstrates that graph models can develop emergent capabilities from scratch, potentially influencing future AI architectures.

RANK_REASON The cluster describes a new academic paper detailing a novel graph foundation model. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• Acacia
• alphaXiv
• arXiv
• CatalyzeX
• Common Crawl
• CORE Recommender
• DagsHub
• Gotit.pub
• Hugging Face
• Influence Flower
• ScienceCast

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →