Retrieve-then-Adapt: Retrieval-Augmented Test-Time Adaptation for Sequential Recommendation

TL;DR

This paper introduces Retrieve-then-Adapt (ReAd), a novel framework for test-time adaptation in sequential recommendation that dynamically incorporates retrieved user preferences to improve prediction accuracy.

Contribution

ReAd is the first framework to effectively combine retrieval-based augmentation with efficient test-time adaptation for sequential recommendation.

Findings

ReAd outperforms existing methods across five benchmark datasets.

The retrieval-augmented approach improves adaptation to real-time user preferences.

ReAd achieves better recommendation accuracy with lower computational overhead.

Abstract

The sequential recommendation (SR) task aims to predict the next item based on users' historical interaction sequences. Typically trained on historical data, SR models often struggle to adapt to real-time preference shifts during inference due to challenges posed by distributional divergence and parameterized constraints. Existing approaches to address this issue include test-time training, test-time augmentation, and retrieval-augmented fine-tuning. However, these methods either introduce significant computational overhead, rely on random augmentation strategies, or require a carefully designed two-stage training paradigm. In this paper, we argue that the key to effective test-time adaptation lies in achieving both effective augmentation and efficient adaptation. To this end, we propose Retrieve-then-Adapt (ReAd), a novel framework that dynamically adapts a deployed SR model to the test distribution through retrieved user preference signals. Specifically, given a trained SR model, ReAd first retrieves collaboratively similar items for a test user from a constructed collaborative memory database. A lightweight retrieval learning module then integrates these items into an informative augmentation embedding that captures both collaborative signals and prediction-refinement cues. Finally, the initial SR prediction is refined via a fusion mechanism that incorporates this embedding. Extensive experiments across five benchmark datasets demonstrate that ReAd consistently outperforms existing SR methods.