Research map / Learning theory and algorithms / Prediction and decision losses
On Component Interactions in Two-Stage Recommender Systems
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Learn item-pool allocation with a Mixture-of-Experts gate instead of using random fixed pools | Two-stage recommender with two-tower nominators · AmazonCat-13K, offline test set | precision@5 | Helped |
| Learn item-pool allocation with a Mixture-of-Experts gate instead of using random fixed pools | Two-stage recommender with two-tower nominators · AmazonCat-13K, offline test set | recall@5 | Helped |
| Use learned pool allocation with the smallest training set size | Mixture-of-Experts pool allocation · AmazonCat-13K, c = 100 samples per arm | precision@5 | Mixed |
| Increase the learned item embedding dimension | Mixture-of-Experts pool allocation · AmazonCat-13K, high item embedding dimension d_e | precision@5 | Mixed |
| Train nominators independently with training-on-own | Two-stage recommender · Constructed supervised and bandit settings | cumulative regret | Hurt |
| Train nominators independently with training-on-all | Two-stage recommender · Constructed supervised and bandit settings | cumulative regret | Hurt |
| Increase the number of nominators | Two-stage recommender · AmazonCat-13K bandit experiments | regret | Helped |
Closest papers by meaning
- Generalization Error Bounds for Two-stage Recommender Systems with Tree Structure (NeurIPS 2024 · similarity 0.88)
- Building Higher-Order Abstractions from the Components of Recommender Systems (AAAI 2024 · similarity 0.87)
- RecFlow: An Industrial Full Flow Recommendation Dataset (ICLR 2025 · similarity 0.86)
- User-Creator Feature Polarization in Recommender Systems with Dual Influence (NeurIPS 2024 · similarity 0.86)
- ORBIT - Open Recommendation Benchmark for Reproducible Research with Hidden Tests (NeurIPS 2025 · similarity 0.85)
- Would You Like Your Data to Be Trained? A User Controllable Recommendation Framework (AAAI 2024 · 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.