cs.LGNov 1, 2026

Search-Aware On-Policy Self-Distillation with Sibling Alignment for Generative Recommendation

Authors: Rongfeng Guo, Zhuang Zhuang, Ji Dai, Peng Zhao, Shijie Li, Jie Chen, Teng Long, Vincent Tao Hu, 

Organizations: Kuaishou Technology, Beijing, China · University of Amsterdam, Amsterdam, Netherlands · Huazhong University of Science and Technology, Wuhan, China

Abstract

Generative recommendation retrieves items by autoregressively generating semantic IDs, typically via beam search. On-policy self-distillation (OPSD) enhances training supervision by transferring privileged self-teacher knowledge at student-generated states. However, beam-search retention depends on competition among prefixes: visited states differ in importance, and initially reliable decisions may approach the pruning boundary as training evolves. Moreover, conventional distribution matching couples target-token confidence with preferences among competing continuations, potentially weakening relational supervision when the self-teacher strongly favors the target. To address these challenges, we propose Search-Aware On-Policy Self-Distillation (SA-OPSD), a framework that transfers privileged self-teacher knowledge to the states visited during the student's own beam search. Specifically, we introduce Evolving-Boundary Sampling, which prioritizes search-critical states by considering both current pruning-boundary proximity and initial-reference retention-margin decline under a controlled distillation budget. Furthermore, we develop Confidence-Decoupled Branch Distillation, which separates target-probability calibration from informative sibling relations to better preserve the teacher's preferences among competing branches. Experiments on Industrial, Office, and Instruments demonstrate strong performance for SA-OPSD.

Figures & tables

Appendix figures & tables2 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. RPSD: Retrieval-Privileged On-Policy Self-Distillation with Optimal Transport

    Oct 27, 2026Dongyue Li, Zebang Cheng, Zhongyi Deng +3Sequential RecommendationRL for Recommendation

  2. SR-OPSD: Self-Referenced On-Policy Self-Distillation

    Aug 10, 2026Zhuo Sun, Entong Li, Yanlong Zhao +9On-Policy Self-DistillationLanguage Model Distillation

  3. Preference-Based Self-Distillation: Beyond KL Matching via Reward Regularization

    May 6, 2026Xin Yu, Liuchen Liao, Yiwen Zhang +3On-Policy Self-DistillationLanguage Model Distillation