Research map / Label-efficient and robust learning / Recognition and long-tail
WalkRanker: A Unified Pairwise Ranking Model With Multiple Relations for Item Recommendation
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 |
|---|---|---|---|
| Add random-walk-based multiple relations (probability relevance) over RankMBPR | pairwise ranking · Yelp-80 | MAP@10: 0.0282 -> 0.0291 | Helped |
| Add random-walk-based multiple relations over RankMBPR | pairwise ranking · Epinions-50 | MAP@10: 0.0242 -> 0.0265 | Helped |
| Add random-walk-based multiple relations over RankMBPR, cold-start users | pairwise ranking · Epinions-50 cold-start users | MAP@10: 0.0168 -> 0.0189 | Helped |
| Use rank-aware weighting (RankMBPR, WalkRanker) instead of AdaBPR dynamic sampling | pairwise ranking · Yelp and Epinions | MAP@10: 0.0252 -> 0.0282 (Yelp-80, AdaBPR -> RankMBPR) | Helped |
| Use adaptive negative sampling (AdaBPR, WARP, RankMBPR) instead of uniform BPR | pairwise ranking · Yelp-80 | MAP@10: 0.0241 -> 0.0282 | Helped |
| Increase random walk length tau (ablation) | WalkRanker · Yelp / Epinions | MAP@10 | Mixed |
| Use unequal weights for context vertices (beta=0.4) instead of equal (beta=0) (ablation) | WalkRanker · Yelp/Epinions | performance | Helped |
| Increase negative subset size n (ablation) | WalkRanker · Yelp/Epinions | performance | Mixed |
Closest papers by meaning
- From Pairwise to Ranking: Climbing the Ladder to Ideal Collaborative Filtering with Pseudo-Ranking (AAAI 2025 · similarity 0.90)
- Preference Diffusion for Recommendation (ICLR 2025 · similarity 0.87)
- Learning-Efficient Yet Generalizable Collaborative Filtering for Item Recommendation (ICML 2024 · similarity 0.87)
- Fading to Grow: Growing Preference Ratios via Preference Fading Discrete Diffusion for Recommendation (NeurIPS 2025 · similarity 0.87)
- On Softmax Direct Preference Optimization for Recommendation (NeurIPS 2024 · similarity 0.87)
- Temporally and Distributionally Robust Optimization for Cold-Start Recommendation (AAAI 2024 · similarity 0.86)
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.