AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Can Recommender Systems Teach Themselves? A Recursive Self-Improving Framework with Fidelity Control

Luankang Zhang, Hao Wang, Zhongzhou Liu, Mingjia Yin, Yonghao Huang, Jiaqi Li, Wei Guo, Yong Liu, Huifeng Guo, Defu Lian, Enhong Chen

ICML 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 fidelity-based quality control (ablation)RSIR (SASRec) · amazon-sport, iteration 1NDCG@10: 0.0273 -> 0.0293Improved
add fidelity-based quality control (ablation)RSIR (SASRec) · amazon-sport, iteration 2NDCG@10: 0.0209 -> 0.0294Improved
add fidelity-based quality control (ablation)RSIR (SASRec) · amazon-sport, iteration 3Recall@10: 0.0210 -> 0.0528Improved
add fidelity-based quality control (ablation)RSIR (SASRec) · amazon-sport, iteration 1Recall@10: 0.0472 -> 0.0512Improved
apply one RSIR iteration to SASRecSASRec · amazon-beautyNDCG@10: 0.0290 -> 0.0322Improved
apply RSIR to HSTUHSTU · amazon-toysRecall@10: 0.0869 -> 0.0924Improved
add RSIR instead of DR4SR augmentationSASRec · amazon-sportRecall@10: 0.0495 -> 0.0512Improved
run additional RSIR iterations (1 to 3) (ablation)RSIR (SASRec) · amazon-sportRecall@10: 0.0512 -> 0.0528Improved
run RSIR for 8 iterations vs 3 (ablation)RSIR (SASRec) · amazon-sportRecall@10: 0.0528 -> 0.0513Hurt
fidelity threshold tau=1 (too strict) (ablation)RSIR · amazon-sportRecall@10: 0.0512 -> 0.0469Hurt
exploitation probability p=1 vs p=0.5 (ablation)RSIR · amazon-sportRecall@10: 0.0512 -> 0.0478Hurt
exploitation probability p=0 vs p=0.5 (ablation)RSIR · amazon-sportNDCG@10: 0.0293 -> 0.0281Hurt

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