A learnable graph token list that adaptively reweights hops and selects informative within-hop nodes alleviates the hop-overpriority problem in tokenized graph learning models, especially on heterophilic graphs.
Less is more: on the over-globalizing problem in graph transformers
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
-
Adaptive Tokenization: On the Hop-Overpriority Problem in Tokenized Graph Learning Models
A learnable graph token list that adaptively reweights hops and selects informative within-hop nodes alleviates the hop-overpriority problem in tokenized graph learning models, especially on heterophilic graphs.