Jun 2026
TAG-DLM: Diffusion Language Models for Text-Attributed Graph Learning
This work proposes method that unifies textual reasoning and graph message passing within a masked diffusion language model, a language model with bidirectional attention and generative decoding that outperforms graph neural networks, graph transformers, and LLM-based baselines on all three TAG benchmarks across two tasks.
Lingjie Chen, Yuanchen Bei, Haobo Xu et al.
· arXiv.org · 1 citation