TIDFormer, a dynamic graph Transformer with interaction-level self-attention and three encoding modules, achieves the best or second-best average rank on seven CTDG link prediction benchmarks.
Title resolution pending
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
-
TIDFormer: Exploiting Temporal and Interactive Dynamics Makes A Great Dynamic Graph Transformer
TIDFormer, a dynamic graph Transformer with interaction-level self-attention and three encoding modules, achieves the best or second-best average rank on seven CTDG link prediction benchmarks.