Frequency-aware Adaptive Contrastive Learning for Sequential Recommendation

2026-01-22

AI Summary

This work addresses the limitations of contrastive learning in sequential recommendation, particularly its inadequate modeling of low-frequency items and sparse user behaviors due to conventional data augmentation strategies. To overcome this, the authors propose a Frequency-Aware Contrastive Learning (FACL) framework that operates at both micro and macro levels. At the micro level, FACL introduces adaptive perturbations to preserve the semantic integrity of low-frequency items; at the macro level, it employs a frequency-aware reweighting strategy to amplify the learning signal from sparse interaction sequences. Extensive experiments on five public datasets demonstrate that FACL significantly outperforms state-of-the-art methods, achieving up to a 3.8% improvement in recommendation accuracy and substantially enhancing performance for long-tail users and items.

Technology Category

Machine Learning: Learning Preferences or RankingsData Mining & Knowledge Management: Recommender SystemsSearch and Optimization: Learning to Search

Application Category

User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systemsSearch and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for rankingSemantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

In this paper, we revisited the role of data augmentation in contrastive learning for sequential recommendation, revealing its inherent bias against low-frequency items and sparse user behaviors. To address this limitation, we proposed FACL, a frequency-aware adaptive contrastive learning framework that introduces micro-level adaptive perturbation to protect the integrity of rare items, as well as macro-level reweighting to amplify the influence of sparse and rare-interaction sequences during training. Comprehensive experiments on five public benchmark datasets demonstrated that FACL consistently outperforms state-of-the-art data augmentation and model augmentation-based methods, achieving up to 3.8% improvement in recommendation accuracy. Moreover, fine-grained analyses confirm that FACL significantly alleviates the performance drop on low-frequency items and users, highlighting its robust intent-preserving ability and its superior applicability to real-world, long-tail recommendation scenarios.

Problem

• sequential recommendation
• contrastive learning
• data augmentation
• low-frequency items
• long-tail scenarios

Innovation

• frequency-aware
• adaptive contrastive learning
• sequential recommendation
• data augmentation
• long-tail recommendation