Test-Time Alignment for Tracking User Interest Shifts in Sequential Recommendation

2025-04-02

AI Summary

To address the degradation of online inference performance in sequential recommendation caused by dynamic user interest drift, this paper proposes T²ARec—a test-time unsupervised adaptive learning framework. Methodologically, it introduces a dual-alignment mechanism: (1) alignment between absolute temporal intervals and model update steps to explicitly encode real-world time scales; and (2) explicit alignment of user interest states, enabling interpretable detection and correction of interest drift. Built upon a state-space modeling foundation, T²ARec integrates time-aware representation learning, self-supervised parameter adaptation, and distribution shift detection. Evaluated on three benchmark datasets, it achieves state-of-the-art performance, significantly improving recommendation accuracy, timeliness, and robustness. The approach establishes a theoretically grounded and empirically effective test-time learning paradigm for modeling dynamically evolving user interests.

Technology Category

• Machine Learning: Time-Series/Data Streams
• Search and Optimization: Learning to Search
• Reasoning under Uncertainty: Sequential Decision Making

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Sequential recommendation is essential in modern recommender systems, aiming to predict the next item a user may interact with based on their historical behaviors. However, real-world scenarios are often dynamic and subject to shifts in user interests. Conventional sequential recommendation models are typically trained on static historical data, limiting their ability to adapt to such shifts and resulting in significant performance degradation during testing. Recently, Test-Time Training (TTT) has emerged as a promising paradigm, enabling pre-trained models to dynamically adapt to test data by leveraging unlabeled examples during testing. However, applying TTT to effectively track and address user interest shifts in recommender systems remains an open and challenging problem. Key challenges include how to capture temporal information effectively and explicitly identifying shifts in user interests during the testing phase. To address these issues, we propose T$^2$ARec, a novel model leveraging state space model for TTT by introducing two Test-Time Alignment modules tailored for sequential recommendation, effectively capturing the distribution shifts in user interest patterns over time. Specifically, T$^2$ARec aligns absolute time intervals with model-adaptive learning intervals to capture temporal dynamics and introduce an interest state alignment mechanism to effectively and explicitly identify the user interest shifts with theoretical guarantees. These two alignment modules enable efficient and incremental updates to model parameters in a self-supervised manner during testing, enhancing predictions for online recommendation. Extensive evaluations on three benchmark datasets demonstrate that T$^2$ARec achieves state-of-the-art performance and robustly mitigates the challenges posed by user interest shifts.

Problem

• Adapting to dynamic user interest shifts in recommendations
• Capturing temporal dynamics during test-time adaptation
• Identifying explicit user interest shifts with guarantees

Innovation

• State space model for Test-Time Training
• Absolute time intervals alignment mechanism
• Interest state alignment with theoretical guarantees