TGFormer is a temporal graph Transformer using an auto-correlation mechanism to uncover periodic dependencies, reporting up to 9.35% precision gains over prior methods on six benchmarks.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
TGFormer: Towards Temporal Graph Transformer with Auto-Correlation Mechanism
TGFormer is a temporal graph Transformer using an auto-correlation mechanism to uncover periodic dependencies, reporting up to 9.35% precision gains over prior methods on six benchmarks.