发表机构
School of Big data and Software Engineering, Chongqing University（重庆大学大数据与软件工程学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究针对基于会话推荐中传统卷积模型全局建模弱的问题，提出纯卷积框架NextConvRec，结合结构位置编码与深浅卷积扩大感受野，在4个基准上性能超SOTA基线约1.73%且推理提速16.7%。
AI 中文摘要
基于会话的推荐（SBR）通过分析近期交互预测会话中的下一个选择。基于Transformer的模型因能通过自注意力机制捕获长程依赖而被广泛使用。相比之下，传统卷积模型尽管效率更高，但通常受限于较弱的全局建模能力，在SBR任务中逐渐失势。本研究提出了面向SBR任务的下一代纯卷积框架（NextConvRec），旨在平衡效率与性能。NextConvRec采用结构与位置卷积编码器（SPCE）进行预处理，将可学习的卷积位置偏差与通过GCN层提取的会话级结构信号相结合。其骨干卷积模块通过深度卷积和逐点卷积有效扩大有效感受野，无需注意力机制即可实现鲁棒的长程偏好建模。在4个基准数据集上的大量实验表明，NextConvRec平均优于多个当前最优基线约1.73%，并将每个会话的平均推理时间降低16.7%。卷积架构仍是实现高效准确的基于会话推荐的一个有前景的方向。
英文摘要
Session-based recommendation (SBR) predicts the next choice in a session by analyzing recent interactions. Transformer-based models are widely used because of their ability to capture long-range dependencies through self-attention mechanisms. In contrast, traditional convolutional models, although more efficient, are often limited by their weak global modeling capabilities and are losing ground in SBR tasks. In this work, we propose a Next-generation Pure Convolutional Framework (NextConvRec) for SBR tasks, aiming to balance efficiency and performance. NextConvRec uses a Structural and Positional Convolutional Encoder (SPCE) for preprocessing, combining learnable convolutional positional biases with session-level structural signals extracted through GCN layers. Its backbone convolutional module effectively expands the effective receptive field through depthwise convolutions and pointwise convolutions, enabling robust long-range preference modeling without attention mechanisms. Extensive experiments on 4 benchmark datasets show that NextConvRec outperforms several state-of-the-art baselines by around 1.73% on average, and reduces the average inference time per session by 16.7%. The convolutional architectures remain a promising direction for efficient and accurate session-based recommendations.
