SSD4Rec: A Structured State Space Duality Model for Efficient Sequential Recommendation

TL;DR

SSD4Rec introduces a novel, efficient sequential recommendation model leveraging structured state space duality, enabling superior long-sequence modeling and scalability, outperforming existing methods on benchmark datasets.

Contribution

The paper adapts the Mamba architecture with structured state space duality for sequential recommendation, achieving state-of-the-art results with efficient long-sequence handling.

Findings

Achieves state-of-the-art performance on four benchmark datasets.

Maintains near-linear scalability with sequence length.

Effectively models variable-length and long-range user behavior sequences.

Abstract

Sequential recommendation methods are crucial in modern recommender systems for their remarkable capability to understand a user's changing interests based on past interactions. However, a significant challenge faced by current methods (e.g., RNN- or Transformer-based models) is to effectively and efficiently capture users' preferences by modeling long behavior sequences, which impedes their various applications like short video platforms where user interactions are numerous. Recently, an emerging architecture named Mamba, built on state space models (SSM) with efficient hardware-aware designs, has showcased the tremendous potential for sequence modeling, presenting a compelling avenue for addressing the challenge effectively. Inspired by this, we propose a novel generic and efficient sequential recommendation backbone, SSD4Rec, which explores the seamless adaptation of Mamba for sequential recommendations. Specifically, SSD4Rec marks the variable- and long-length item sequences with sequence registers and processes the item representations with bidirectional Structured State Space Duality (SSD) blocks. This not only allows for hardware-aware matrix multiplication but also empowers outstanding capabilities in variable-length and long-range sequence modeling. Extensive evaluations on four benchmark datasets demonstrate that the proposed model achieves state-of-the-art performance while maintaining near-linear scalability with user sequence length. Our code is publicly available at https://github.com/ZhangYifeng1995/SSD4Rec.