Exploiting Inter-Session Information with Frequency-enhanced Dual-Path Networks for Sequential Recommendation

TL;DR

FreqRec is a novel sequential recommendation model that leverages inter-session spectral dependencies and frequency-domain alignment to improve prediction accuracy, especially in sparse and noisy data scenarios.

Contribution

It introduces a frequency-enhanced dual-path network that jointly models inter- and intra-session behaviors with spectral alignment, addressing limitations of existing frequency-aware methods.

Findings

Outperforms strong baselines on three benchmark datasets.

Maintains robustness under data sparsity and noisy logs.

Effectively captures cross-session spectral dependencies.

Abstract

Sequential recommendation (SR) aims to predict a user's next item preference by modeling historical interaction sequences. Recent advances often integrate frequency-domain modules to compensate for self-attention's low-pass nature by restoring the high-frequency signals critical for personalized recommendations. Nevertheless, existing frequency-aware solutions process each session in isolation and optimize exclusively with time-domain objectives. Consequently, they overlook cross-session spectral dependencies and fail to enforce alignment between predicted and actual spectral signatures, leaving valuable frequency information under-exploited. To this end, we propose FreqRec, a Frequency-Enhanced Dual-Path Network for sequential Recommendation that jointly captures inter-session and intra-session behaviors via a learnable Frequency-domain Multi-layer Perceptrons. Moreover, FreqRec is optimized under a composite objective that combines cross entropy with a frequency-domain consistency loss, explicitly aligning predicted and true spectral signatures. Extensive experiments on three benchmarks show that FreqRec surpasses strong baselines and remains robust under data sparsity and noisy-log conditions.