AI RESEARCH
Graph-Aware Text-Only Backdoor Poisoning for Text-Attributed Graphs
arXiv CS.LG
•
ArXi:2603.20339v1 Announce Type: new Many learning systems now use graph data in which each node also contains text, such as papers with abstracts or users with posts. Because these texts often come from open platforms, an attacker may be able to quietly poison a small part of the