Research map / Graph learning / Knowledge and temporal graphs
Revisting Node Affinity Prediction In Temporal Graphs
Or Feldman, Krishna Sri Ipsit Mantri, Moshe Eliasof, Chaim Baskin
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace GRU state update with linear SSM update (and add global vector) (ablation) | NAViS · tgbn-genre | NDCG@10: 0.398 -> 0.528 | Improved |
| add global virtual vector (ablation) | NAViS · tgbn-genre | NDCG@10: 0.521 -> 0.528 | Improved |
| add global virtual vector (ablation) | NAViS · tgbn-token | NDCG@10: 0.511 -> 0.513 | Improved |
| use rank loss instead of cross-entropy (ablation) | NAViS · tgbn-genre | NDCG@10: 0.461 -> 0.528 | Improved |
| use rank loss instead of cross-entropy (ablation) | NAViS · tgbn-trade | NDCG@10: 0.859 -> 0.863 | Improved |
| add pairwise margin regularization (ablation) | NAViS · tgbn-reddit | NDCG@10: 0.553 -> 0.569 | Improved |
| change margin delta (0.1/0.01/0.001) or batch size (ablation) | NAViS · tgbn-trade, tgbn-genre | NDCG@10: 0.863 for all; genre 0.526-0.528 | No clear effect |
| change global buffer size or aggregation scheme (ablation) | NAViS · tgbn-genre | NDCG@10: 0.528 -> 0.527 at worst | No clear effect |
| NAViS vs S4 block | NAViS · tgbn-trade | NDCG@10: 0.796 (S4) -> 0.863 | Improved |
| NAViS vs best TGNN (TGNv2) | NAViS · Wikipedia/Flights/USLegis/UNVote | NDCG@10: 0.813 -> 0.952 (UNVote) | Improved |
| NAViS vs Moving Avg in previous-affinity setting | NAViS · tgbn-trade | NDCG@10: 0.855 (PF) vs 0.863 | Mixed |
Closest papers by meaning
- Future Link Prediction Without Memory or Aggregation (NeurIPS 2025 · similarity 0.87)
- Using Time-Aware Graph Neural Networks to Predict Temporal Centralities in Dynamic Graphs (NeurIPS 2024 · similarity 0.87)
- A Topological Perspective on Demystifying GNN-Based Link Prediction Performance (ICLR 2024 · similarity 0.86)
- Improving Temporal Link Prediction via Temporal Walk Matrix Projection (NeurIPS 2024 · similarity 0.86)
- Spear and Shield: Adversarial Attacks and Defense Methods for Model-Based Link Prediction on Continuous-Time Dynamic Graphs (AAAI 2024 · similarity 0.85)
- TAMI: Taming Heterogeneity in Temporal Interactions for Temporal Graph Link Prediction (NeurIPS 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
