Search papers, labs, and topics across Lattice.
The Education University of Hong Kong
2
0
4
0
GNN performance on heterophilic graphs suffers because of inductive subgraphs acting as spurious shortcuts, a problem that can be solved by causally disentangling these subgraphs.
By dynamically weighting historical interactions, TIPS lets sequential recommenders see past the biases of what users *actually* clicked, revealing what they *would* have clicked.
