Research map / Label-efficient and robust learning / Recognition and long-tail
CAFU: Constrained Alignment and Filtered Uniformity for Denoising Recommendation
Xinzhe Jiang, Lei Sang, Yi Zhang, Kaibin Wang, Yiwen Zhang
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 constrained (top-N) alignment (ablation) | DirectAU+filtered uniformity · Amazon-Book | Recall@20: 0.0553 -> 0.0565 | Improved |
| add constrained (top-N) alignment (ablation) | DirectAU+filtered uniformity · Tmall | Recall@20: 0.0894 -> 0.0923 | Improved |
| add constrained (top-N) alignment (ablation) | DirectAU+filtered uniformity · Yelp2018 | Recall@20: 0.0737 -> 0.0746 | Improved |
| add filtered uniformity (ablation) | DirectAU+constrained alignment · Amazon-Book | Recall@20: 0.0550 -> 0.0565 | Improved |
| add filtered uniformity (ablation) | DirectAU+constrained alignment · Tmall | Recall@20: 0.0905 -> 0.0923 | Improved |
| add filtered uniformity (ablation) | DirectAU+constrained alignment · Yelp2018 | Recall@20: 0.0735 -> 0.0746 | Improved |
| apply CAFU to DirectAU | DirectAU · Tmall | Recall@20: 0.0835 -> 0.0923 | Improved |
| apply CAFU to MAWU | MAWU · Amazon-Book | Recall@20: 0.0526 -> 0.0568 | Improved |
| apply CAFU to SIURec | SIURec · Yelp2018 | Recall@20: 0.0641 -> 0.0661 | Improved |
| apply DeCA denoising to DirectAU | DirectAU · Amazon-Book | Recall@20: 0.0508 -> 0.0518 | No clear effect |
| apply DeCA denoising to DirectAU (ML-100K) | DirectAU · ML-100K | Recall@20: 0.1209 -> 0.1129 | Hurt |
| apply DeCA denoising to DirectAU (Modcloth) | DirectAU · Modcloth | Recall@20: 0.1837 -> 0.1506 | Hurt |
Closest papers by meaning
- Sub-Interest-Aware Representation Uniformity for Recommender System (AAAI 2025 · similarity 0.90)
- On Efficiency-Effectiveness Trade-off of Diffusion-based Recommenders (NeurIPS 2025 · similarity 0.86)
- FACE: A General Framework for Mapping Collaborative Filtering Embeddings into LLM Tokens (NeurIPS 2025 · similarity 0.86)
- Semantic Convergence: Harmonizing Recommender Systems via Two-Stage Alignment and Behavioral Semantic Tokenization (AAAI 2025 · similarity 0.84)
- ImplicitSLIM and How it Improves Embedding-based Collaborative Filtering (ICLR 2024 · similarity 0.84)
- Relaxing the Accurate Imputation Assumption in Doubly Robust Learning for Debiased Collaborative Filtering (ICML 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.
