there has been recent advances in Positional Enconding for MP-GNNs, such as Laplacian Positional Enconding (LapPE) or Random Walk Positional Encoding (RWSE). However, our smiles graph doesn't take advantage of that, altough recent benchmarks shows that GNN benefit fro m injecting PE informations in GNN [Grötschla et al., 2024]
there has been recent advances in Positional Enconding for MP-GNNs, such as Laplacian Positional Enconding (LapPE) or Random Walk Positional Encoding (RWSE). However, our smiles graph doesn't take advantage of that, altough recent benchmarks shows that GNN benefit fro m injecting PE informations in GNN [Grötschla et al., 2024]