AI RESEARCH

Can LLMs Fool Graph Learning? Exploring Universal Adversarial Attacks on Text-Attributed Graphs

arXiv CS.AI

ArXi:2603.21155v1 Announce Type: new Text-attributed graphs (