Efficient Dataset Selection for Continual Adaptation of Generative Recommenders

TL;DR

This paper explores efficient data selection methods to improve continual adaptation of recommendation systems, focusing on targeted sampling strategies that enhance performance while reducing training costs.

Contribution

It introduces gradient-based representations combined with distribution-matching techniques for effective data curation in scalable recommendation system updates.

Findings

Gradient-based representations improve model performance.

Distribution-matching enhances data selection effectiveness.

Targeted data selection maintains robustness to user behavior drift.

Abstract

Recommendation systems must continuously adapt to evolving user behavior, yet the volume of data generated in large-scale streaming environments makes frequent full retraining impractical. This work investigates how targeted data selection can mitigate performance degradation caused by temporal distributional drift while maintaining scalability. We evaluate a range of representation choices and sampling strategies for curating small but informative subsets of user interaction data. Our results demonstrate that gradient-based representations, coupled with distribution-matching, improve downstream model performance, achieving training efficiency gains while preserving robustness to drift. These findings highlight data curation as a practical mechanism for scalable monitoring and adaptive model updates in production-scale recommendation systems.