Lighter-X: An Efficient and Plug-and-play Strategy for Graph-based Recommendation through Decoupled Propagation
Summary: Introduces Lighter‑X: a plug‑and‑play compression for GNN recommenders reducing parameters from O(n×d) to O(h×d) (h << n) via compressed adjacency and embeddings while keeping theoretical guarantees. Decoupled propagation cuts training cost and matches/outperforms LightGCN on million-edge graphs with ~1% parameters. (summarized by gpt-5-mini on Feb 09 2026)
Incoming Non-self Citations Over Time
No non-self incoming citations found for this paper in this database.
Authors
- 1. Yanping Zheng
- 2. Zhewei Wei
- 3. Frank de Hoog
- 4. Xu Chen
- 5. Hongteng Xu
- 6. Yuhang Ye
- 7. Jiadeng Huang
Incoming Citations (Sorted by Pagerank)
Showing 0 of 0 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 2 of 2 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 2,688 | Accelerating Recommendation System Training by Leveraging Popular Choices | 2022 | VLDB | 8.2991144e-05 |
| 5,147 | Efficient Tree-SVD for Subset Node Embedding over Large Dynamic Graphs | 2023 | SIGMOD | 5.6643767e-05 |
Previous
Page 1 / 1
Next