作者位次 #4/8SIGIR 2026Oral· Information Retrieval (Full)
Learning Noise-Resilient and Transferable Graph-Text Alignment via Dynamic Quality Assessment
Yuhang Liu · Minglai Shao · Zengyi Wo · Yunlong Chu · Bing Hao · Shengzhong Liu · Ruijie Wang · Jianxin Li
摘要
Pre-training graph foundation models (GFMs) on text-attributed graphs (TAGs) is important for web-scale retrieval and recommendation, where graph entities are matched with textual descriptions. Existing CLIP-style graph-text aligners typically assume one-to-one correspondence: each node is pulled...