Research map / Label-efficient and robust learning / Recognition and long-tail
Incorporating Bias-aware Margins into Contrastive Loss for Collaborative Filtering
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 |
|---|---|---|---|
| train with BC loss instead of softmax-free MF baseline loss | MF · Tencent imbalanced | Recall@20: 0.0982 -> 0.1298 | Helped |
| train with BC loss | LightGCN · Amazon-book balanced | Recall@20: 0.0123 -> 0.0257 | Helped |
| train with BC loss | MF · Douban temporal split | Recall: 0.0294 -> 0.0324 | Helped |
| use MACR debiasing | MF · Tencent imbalanced overall | NDCG@20: 0.06431 -> 0.02413 | Hurt |
| use IPS-CN debiasing | MF · Tencent imbalanced | Recall@20: 0.0982 -> 0.0686 | Hurt |
| use IPS-CN debiasing | MF · Tencent balanced | Recall@20: 0.0052 -> 0.0075 | Helped |
| use BC loss on head items | MF · Tencent imbalanced head | NDCG@20: 0.06837 -> 0.09552 | Helped |
| use BC loss on tail items | MF · Tencent balanced tail | NDCG@20: 0.00004 -> 0.00024 | Helped |
| use adaptive bias-aware margin instead of constant margin (ablation) | BC loss · Tencent | Recall/NDCG | Helped |
| add margin to softmax loss (BC loss vs softmax) (ablation) | softmax loss · Tencent | Recall/NDCG | Helped |
| vary temperature tau (ablation) | BC loss · Tencent | Recall/NDCG | Mixed |
Closest papers by meaning
- Learning-Efficient Yet Generalizable Collaborative Filtering for Item Recommendation (ICML 2024 · similarity 0.87)
- Relaxing the Accurate Imputation Assumption in Doubly Robust Learning for Debiased Collaborative Filtering (ICML 2024 · similarity 0.87)
- From Pairwise to Ranking: Climbing the Ladder to Ideal Collaborative Filtering with Pseudo-Ranking (AAAI 2025 · similarity 0.87)
- Debiased Collaborative Filtering with Kernel-Based Causal Balancing (ICLR 2024 · similarity 0.87)
- Advancing Loss Functions in Recommender Systems: A Comparative Study with a Rényi Divergence-Based Solution (AAAI 2025 · similarity 0.87)
- Alleviating Dual Biases in Recommendation (Student Abstract) (AAAI 2025 · 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.