Papers › Weisfeiler and Lehman Go Cellular: CW Networks

Weisfeiler and Lehman Go Cellular: CW Networks

23 Jun 2021NeurIPS 2021 12arXiv:2106.12575archive 2025-07-28

Cristian Bodnar, Fabrizio Frasca, Nina Otter, Yu Guang Wang, Pietro Liò, Guido Montúfar, Michael Bronstein

Graph Neural Networks (GNNs) are limited in their expressive power, struggle with long-range interactions and lack a principled way to model higher-order structures. These problems can be attributed to the strong coupling between the computational graph and the input graph structure. The recently proposed Message Passing Simplicial Networks naturally decouple these elements by performing message passing on the clique complex of the graph. Nevertheless, these models can be severely constrained by the rigid combinatorial structure of Simplicial Complexes (SCs). In this work, we extend recent theoretical results on SCs to regular Cell Complexes, topological objects that flexibly subsume SCs and graphs. We show that this generalisation provides a powerful set of graph "lifting" transformations, each leading to a unique hierarchical message passing procedure. The resulting methods, which we collectively call CW Networks (CWNs), are strictly more powerful than the WL test and not less powerful than the 3-WL test. In particular, we demonstrate the effectiveness of one such scheme, based on rings, when applied to molecular graph problems. The proposed architecture benefits from provably larger expressivity than commonly used GNNs, principled modelling of higher-order signals and from compressing the distances between nodes. We demonstrate that our model achieves state-of-the-art results on a variety of molecular datasets.

PaperPDFConference PDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

twitter-research/cwn officialmentioned in papermentioned on GitHubpytorchMIT report
josefhoppe/cwn-random-ccs mentioned on GitHubpytorchMIT report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Graph ClassificationGraph Property PredictionGraph Regression

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Graph Classification CSL CIN Acc 1 #1 of 1 Archive leaderboard report
Graph Property Prediction ogbg-molhiv CIN Ext. data No #10 of 43 Archive leaderboard report
Graph Property Prediction ogbg-molhiv CIN Number of params 239745 #10 of 43 Archive leaderboard report
Graph Property Prediction ogbg-molhiv CIN Test ROC-AUC 0.8094 ± 0.0057 #10 of 43 Archive leaderboard report
Graph Property Prediction ogbg-molhiv CIN Validation ROC-AUC 0.8277 ± 0.0099 #10 of 43 Archive leaderboard report
Graph Property Prediction ogbg-molhiv CIN-small Ext. data No #14 of 43 Archive leaderboard report
Graph Property Prediction ogbg-molhiv CIN-small Number of params 138337 #14 of 43 Archive leaderboard report
Graph Property Prediction ogbg-molhiv CIN-small Test ROC-AUC 0.8055 ± 0.0104 #14 of 43 Archive leaderboard report
Graph Property Prediction ogbg-molhiv CIN-small Validation ROC-AUC 0.8310 ± 0.0102 #14 of 43 Archive leaderboard report
Graph Regression ZINC CIN MAE 0.079 #16 of 27 Archive leaderboard report
Graph Regression ZINC CIN-small MAE 0.094 #19 of 27 Archive leaderboard report
Graph Regression ZINC 100k CIN-small MAE 0.094 #1 of 8 Archive leaderboard report
Graph Regression ZINC-500k CIN MAE 0.079 #15 of 36 Archive leaderboard report
Graph Regression ZINC-500k CIN-small MAE 0.094 #19 of 36 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections