Efficient Personalized Reranking with Semi-Autoregressive Generation and Online Knowledge Distillation

TL;DR

This paper introduces PSAD, a novel reranking framework that combines semi-autoregressive generation with online knowledge distillation and user profile modeling to improve ranking quality and efficiency in recommender systems.

Contribution

The paper proposes a new framework that balances generation quality and speed, and enhances user-item interaction modeling for personalized reranking.

Findings

PSAD outperforms state-of-the-art methods in ranking accuracy.

PSAD achieves significant improvements in inference efficiency.

The framework effectively models user interest dynamics.

Abstract

Generative models offer a promising paradigm for the final stage reranking in multi-stage recommender systems, with the ability to capture inter-item dependencies within reranked lists. However, their practical deployment still faces two key challenges: (1) an inherent conflict between achieving high generation quality and ensuring low-latency inference, making it difficult to balance the two, and (2) insufficient interaction between user and item features in existing methods. To address these challenges, we propose a novel Personalized Semi-Autoregressive with online knowledge Distillation (PSAD) framework for reranking. In this framework, the teacher model adopts a semi-autoregressive generator to balance generation quality and efficiency, while its ranking knowledge is distilled online into a lightweight scoring network during joint training, enabling real-time and efficient inference. Furthermore, we propose a User Profile Network (UPN) that injects user intent and models interest dynamics, enabling deeper interactions between users and items. Extensive experiments conducted on three large-scale public datasets demonstrate that PSAD significantly outperforms state-of-the-art baselines in both ranking performance and inference efficiency.