Recommendation ranking improves with sparse feedback using pretrained user routed adapters
SAGE-PAIR: Source-Pretrained Routed Residual Adaptation for Sparse-Feedback Recommendation
Information Retrieval
Summary
Recommendation systems often struggle when they start serving new content without enough feedback to update their models. This paper studies very sparse feedback scenarios where existing methods have trouble learning useful updates. The authors propose SAGE-PAIR, a technique that first learns corrections from previous data and then adjusts these corrections using limited new feedback, while also tailoring updates to different user profiles. Their method improves recommendation quality with little extra computation and works better than comparable approaches on a public dataset.
What this means in practice
- •For e-commerce engineers: Adapt recommendation models quickly for new product releases using very limited user feedback while maintaining or improving ranking accuracy.
- •For streaming service developers: Improve ranking of suggested content in early serving periods when feedback is scarce by calibrating pretrained user-specific residual corrections.
Tested on one dataset.
Authors
Weiqing Zeng, Zida Yang, Guquan You, Jiexin Zheng, Cong Wang
Abstract
Recommendation systems often enter a new serving window before enough outcomes are available to retrain the deployed model. We study this regime with 0.05-1.0% target feedback, a range that spans our stress tests while keeping supervision far below the degrees of freedom of the source recommender. At this scale, existing fine-tuning and meta-learning methods still estimate target-specific updates from extremely sparse feedback, leaving the update direction weakly identified. SAGE-PAIR instead transfers a source-supported residual correction space and only calibrates this space using target outcomes. We introduce SAGE-PAIR, a staged adapter that pretrains a low-dimensional global residual on an adjacent source period and adds a soft router over residual experts conditioned on source user profiles. Target feedback then calibrates this residual state rather than relearning the recommender. On a locked test split of KuaiSAR-v2, SAGE-PAIR (Full) achieves a 1.77% relative improvement in NDCG@10, while a core-protocol comparison shows a further 0.25% gain from routed experts beyond source-pretrained PAIR. Matched source-pretrained LoRA and Adapter-Tuning baselines remain near the source model, supporting the need for a different adaptation space. Thus, SAGE-PAIR improves ranking quality with millisecond-scale adaptation and inference costs. Sparse-feedback adaptation is fundamentally governed by the identifiability of the update space rather than by label count alone. Source pretraining supplies estimable correction directions for scarce labels, while user-conditioned routing contributes a consistent, differentiated cross-user correction on top of the global correction. Our code is available at https://anonymous.4open.science/r/SAGE-PAIR-610E/.