AI Paper Atlas

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

AAAI 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Recommendation representation learning · 0 citations (OpenAlex)

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 changedWhereMeasuredEffect
add constrained (top-N) alignment (ablation)DirectAU+filtered uniformity · Amazon-BookRecall@20: 0.0553 -> 0.0565Improved
add constrained (top-N) alignment (ablation)DirectAU+filtered uniformity · TmallRecall@20: 0.0894 -> 0.0923Improved
add constrained (top-N) alignment (ablation)DirectAU+filtered uniformity · Yelp2018Recall@20: 0.0737 -> 0.0746Improved
add filtered uniformity (ablation)DirectAU+constrained alignment · Amazon-BookRecall@20: 0.0550 -> 0.0565Improved
add filtered uniformity (ablation)DirectAU+constrained alignment · TmallRecall@20: 0.0905 -> 0.0923Improved
add filtered uniformity (ablation)DirectAU+constrained alignment · Yelp2018Recall@20: 0.0735 -> 0.0746Improved
apply CAFU to DirectAUDirectAU · TmallRecall@20: 0.0835 -> 0.0923Improved
apply CAFU to MAWUMAWU · Amazon-BookRecall@20: 0.0526 -> 0.0568Improved
apply CAFU to SIURecSIURec · Yelp2018Recall@20: 0.0641 -> 0.0661Improved
apply DeCA denoising to DirectAUDirectAU · Amazon-BookRecall@20: 0.0508 -> 0.0518No clear effect
apply DeCA denoising to DirectAU (ML-100K)DirectAU · ML-100KRecall@20: 0.1209 -> 0.1129Hurt
apply DeCA denoising to DirectAU (Modcloth)DirectAU · ModclothRecall@20: 0.1837 -> 0.1506Hurt

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy