Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Temporal Sheaf Neural Networks with Dynamic Orthogonal Transport

About

We introduce Temporal Sheaf Neural Networks (TSNN), a temporal link prediction framework that equips each node with a time-varying orthogonal frame and compares node states only after explicit transport between local coordinate systems. In contrast to existing continuous-time graph models that operate in a shared global embedding space, TSNN models node-specific and evolving interaction semantics through dynamic local frames. The model parameterizes per-node frames via efficient low-rank Householder products, preserves stored hidden states exactly under frame updates, and uses a geometric-residual decoder that anchors predictions on transported distances while learning residual corrections. All computations are strictly causal and use only the pre-event history. We show that the symmetric degree-normalized sheaf Laplacian is orthogonally similar to the symmetric normalized graph Laplacian, with the random-walk normalized form similar in the corresponding degree metric; the full-active, feature-scaled diffusion used by TSNN is exactly a metric-gradient step on the combinatorial sheaf Dirichlet energy, with a degree-free monotone-descent and non-expansiveness guarantee. Frame drift perturbs updates only linearly. Across TGB v2 link-prediction and temporal-heterogeneous leaderboards, together with the DGB benchmark suite, TSNN matches or surpasses the strongest prior methods on most benchmarks, with the largest improvements on graphs exhibiting strong node-role heterogeneity. Ablations confirm the distinct benefit of dynamic frames, orthogonal transport, and geometric-residual decoding.

Md Sadek Hossain Asif, Tanzila Khan, Md. Mosaddek Khan• 2026

Related benchmarks

TaskDatasetResultRank
Future Link PredictionSocialEvolution inductive
AP97.23
89
Inductive dynamic link predictionMOOC (inductive)
AP99.43
42
Temporal Link PredictionWikipedia (inductive)
AUC-ROC99.1
42
Temporal Link PredictionReddit (inductive)
AUC-ROC99.32
42
Dynamic Link Predictiontgbl-wiki (test)
MRR0.8282
19
Link-property predictiontgbl-review v2 (test)
MRR50.28
18
Temporal Link PredictionWikipedia DGB
AP99.41
13
Temporal Link PredictionReddit DGB
AP99.66
13
Temporal Link PredictionMOOC DGB
AP99.36
13
Temporal Link PredictionLastFM DGB
Average Precision (AP)94.51
13
Showing 10 of 29 rows

Other info

Follow for update