Research map / Graph learning / Knowledge and temporal graphs
Parameter-free Dynamic Graph Embedding for Link Prediction
Jiahao Liu, Dongsheng Li, Hansu Gu, Tun Lu, Peng Zhang, Ning Gu
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 |
|---|---|---|---|
| add frequency-aware preference matrix reconstruction (ablation) | FreeGEM (with pattern modeller) · Video | Recall@10: 0.080 -> 0.113 | Improved |
| add frequency-aware preference matrix reconstruction (ablation) | FreeGEM (with pattern modeller) · ML-100K | Recall@10: 0.113 -> 0.114 | No clear effect |
| add personalized interaction pattern modeller (ablation) | FreeGEM (with reconstruction) · ML-100K | Recall@10: 0.074 -> 0.114 | Improved |
| add attribute-integrated SVD (ablation) | FreeGEM · ML-100K | Recall@10: 0.114 -> 0.149 | Improved |
| add dynamic time decay (ablation) | FreeGEM · Wikipedia | MRR: 0.541 -> 0.786 | Improved |
| add attention (ablation) | FreeGEM · LastFM | MRR: 0.163 -> 0.195 | Improved |
| add attention (ablation) | FreeGEM · Wikipedia | MRR: 0.779 -> 0.786 | No clear effect |
| combine offline SVD with online SVD guided by Monitor (ablation) | FreeGEM · Wikipedia | MRR: 0.510 -> 0.530 | Improved |
| Monitor-triggered restart instead of fixed-time restart (ablation) | FreeGEM · Wikipedia/LastFM | approximation error: Time errors 5.28% / 9.67% higher | Improved |
| FreeGEM vs CoPE | FreeGEM · Wikipedia | Hit@10: 0.890 -> 0.852 | Hurt |
| parameter-free embedding instead of backprop-trained models | FreeGEM · Wikipedia | running time: 350.0min -> 9.7min (vs JODIE) | Improved |
Closest papers by meaning
- Practicable Black-Box Evasion Attacks on Link Prediction in Dynamic Graphs—a Graph Sequential Embedding Method (AAAI 2025 · similarity 0.86)
- Future Link Prediction Without Memory or Aggregation (NeurIPS 2025 · similarity 0.86)
- TAMI: Taming Heterogeneity in Temporal Interactions for Temporal Graph Link Prediction (NeurIPS 2025 · similarity 0.85)
- Revisiting Link Prediction: a data perspective (ICLR 2024 · similarity 0.85)
- Decoupling User Relationships Guides Information Diffusion Prediction (Student Abstract) (AAAI 2024 · similarity 0.84)
- A Deep Probabilistic Framework for Continuous Time Dynamic Graph Generation (AAAI 2025 · similarity 0.84)
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.
