G2PT represents graphs as node-then-edge token sequences and learns them with GPT-style next-token prediction, matching or beating diffusion baselines on seven graph and molecule datasets.
2024; Xu et al., 2024; Wang et al.,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Graph Generative Pre-trained Transformer
G2PT represents graphs as node-then-edge token sequences and learns them with GPT-style next-token prediction, matching or beating diffusion baselines on seven graph and molecule datasets.