← Publications

Bridging Theory and Practice in Link Representation with Graph Neural Networks

Veronica Lachi*1, Francesco Ferrini*2, Antonio Longa2, Bruno Lepri1, Andrea Passerini2, Manfred Jaeger3

1Fondazione Bruno Kessler, Trento, Italy · 2University of Trento, Trento, Italy · 3University of Aalborg, Aalborg, Denmark · *Equal contribution

39th Conference on Neural Information Processing Systems (NeurIPS 2025) · Advances in Neural Information Processing Systems 38

Published conference paper · Main Conference Track

In one sentence. This work unifies message-passing link models by expressiveness, introduces LR-EXP to test whether they distinguish structurally different links, and shows that graph symmetry predicts when expressive models are useful.

Problem

GNN expressiveness theory largely concerns whole-graph representations, while link prediction depends on representations of node pairs. Standard GNNs aggregate endpoint embeddings and can collapse non-automorphic links when their endpoints are automorphic. The paper asks how link models can be compared formally and when additional link-level structural expressiveness improves real-world predictions.

Main contributions

  • The kφ-kρ-m framework characterizes endpoint encoders, link-neighborhood encoders, their expressive power, and accessible radius.
  • Formal results place Pure GNNs, ELPH, Neo-GNN, NCN/NCNC, and SEAL in an expressiveness hierarchy.
  • LR-EXP supplies 1,400 synthetic graphs containing non-automorphic links with 1-WL-indistinguishable endpoints.
  • A graph-symmetry score connects structural ambiguity to model performance across 12 link-prediction datasets.

Framework and evaluation

The framework builds a link representation from endpoint embeddings computed by a message-passing function φ and an aggregated m-hop neighborhood representation computed by ρ. Their Weisfeiler–Lehman expressiveness levels are kφ and kρ. LR-EXP trains siamese copies of each model with contrastive loss and applies a reliable paired comparison to determine whether two non-automorphic links receive distinguishable embeddings. Real-world experiments use an efficient WL-based approximation of graph symmetry.

Experimental setting

Tasks
Link-representation discrimination on LR-EXP and link prediction under the HeaRT evaluation setting.
Datasets
LR-EXP; Cora, CiteSeer, PubMed; ogbl-citation2, ogbl-ddi, ogbl-ppa, ogbl-collab; NSC, YST, GRQ, AIFB, and EDIT-TSW.
Metrics
Precision on LR-EXP; mean reciprocal rank (MRR), Hits@10/20, and graph symmetry on real datasets.
Models
GCN, GAT, GraphSAGE, GAE, BUDDY/ELPH, Neo-GNN, NCN, NCNC, and SEAL.

Key results

On LR-EXP, pure GNNs score zero precision, NCN and Neo-GNN reach 75%, and SEAL reaches 97%, matching the theoretical ordering. On low-symmetry real datasets, simpler models can remain competitive. Across the six most symmetric datasets, no pure GNN ranks in the top three, while SEAL ranks first, supporting symmetry-aware model selection.

Selected results (mean ± standard deviation over five runs; higher is better).
Dataset / taskMetricSEALBest pure-GNN baselineDifference
LR-EXP / link distinctionPrecision97 ± 00 ± 0+97 points
YST / link predictionMRR17.51 ± 0.94GAE: 2.32 ± 0.01+15.19
GRQ / link predictionMRR56.72 ± 1.35GCN: 6.89 ± 0.47+49.83
EDIT-TSW / link predictionMRR25.82 ± 1.47GraphSAGE: 8.35 ± 7.27+17.47

When this work is relevant

This paper is relevant for expressive GNNs for link prediction, pairwise or edge representations, limits of endpoint-only embeddings, automorphic-node failures, structural-feature methods, synthetic expressiveness benchmarks, and choosing between fast pure GNNs and subgraph-based models such as SEAL.

Limitations

The theory covers message-passing link models and does not directly include transformer-based or spectral link architectures. LR-EXP is intentionally small and tests discrimination rather than scalability. The real high-symmetry datasets are also relatively small, so larger benchmarks are needed to study generalization in this regime.

Citation

Persistent identifiers: arXiv:2506.24018 · https://doi.org/10.48550/arXiv.2506.24018

@inproceedings{lachi2025bridging, title = {Bridging Theory and Practice in Link Representation with Graph Neural Networks}, author = {Lachi, Veronica and Ferrini, Francesco and Longa, Antonio and Lepri, Bruno and Passerini, Andrea and Jaeger, Manfred}, booktitle = {Advances in Neural Information Processing Systems}, volume = {38}, year = {2025}, url = {https://papers.nips.cc/paper_files/paper/2025/hash/b1cd72276feff8173462e3f733ac66f8-Abstract-Conference.html} }
← Back to publications