Research map / Graph learning / Knowledge and temporal graphs
Task-Aware Retrieval Augmentation for Dynamic Recommendation
Zhen Tao, Xinke Jiang, Qingshuai Feng, Haoyu Zhang, Lun Du, Yuchen Fang, Hao Miao, Bangquan Xie, Qingqiang Sun
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 task-aware retrieval (TarDGR/FT) over RAGRAPH/FT | GraphPro · Koubei | Recall@20: 34.27 -> 36.52 | Improved |
| add task-aware retrieval (TarDGR/FT) over RAGRAPH/FT | GraphPro · Amazon | Recall@20: 18.69 -> 19.56 | Improved |
| add task-aware retrieval (TarDGR/FT) over RAGRAPH/FT | GraphPro · Taobao | Recall@20: 24.78 -> 25.20 | Improved |
| fine-tune the retrieval-augmented model (NF to FT) | TarDGR · Koubei | Recall@20: 24.83 -> 36.52 | Improved |
| add semantic and structural encoders to task-aware model (ablation) | TarDGR · Koubei | Recall@20: 34.14 -> 36.52 | Improved |
| add semantic encoder (ablation) | TarDGR · Koubei | Recall@20: 35.84 -> 36.52 | Improved |
| add structural encoder (ablation) | TarDGR · Taobao | Recall@20: 25.14 -> 25.20 | Improved |
| increase Top-M retrieved subgraphs beyond optimum (ablation) | TarDGR · Taobao/Koubei | Recall/nDCG | Hurt |
| increase k-hop neighborhood depth beyond optimum (ablation) | TarDGR · Taobao/Koubei | Recall/nDCG | Hurt |
Closest papers by meaning
- Temporal Graph Contrastive Learning for Sequential Recommendation (AAAI 2024 · similarity 0.84)
- Semantic Enhanced Heterogeneous Hypergraph Network for Collaborative Filtering (AAAI 2025 · similarity 0.84)
- LIME: Link-based User-item Interaction Modeling with Decoupled XOR Attention for Efficient Test Time Scaling (NeurIPS 2026 · similarity 0.84)
- RAGraph: A General Retrieval-Augmented Graph Learning Framework (NeurIPS 2024 · similarity 0.84)
- ContextGNN: Beyond Two-Tower Recommendation Systems (ICLR 2025 · similarity 0.84)
- Graph-enhanced Optimizers for Structure-aware Recommendation Embedding Evolution (NeurIPS 2024 · 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.
