A Systematic Reproducibility Study of BSARec for Sequential Recommendation
This work provides incremental validation for BSARec in sequential recommendation, assessing its components and proposing metrics for high-frequency signals, but shows limited novelty in method improvements.
The study reproduced BSARec, a sequential recommendation model that uses a frequency layer to address the Transformer's low-pass filtering issue, and found it outperforms other methods on some datasets, with non-constant padding improving performance by enhancing high-frequency signal capture.
In sequential recommendation (SR), the self-attention mechanism of Transformer-based models acts as a low-pass filter, limiting their ability to capture high-frequency signals that reflect short-term user interests. To overcome this, BSARec augments the Transformer encoder with a frequency layer that rescales high-frequency components using the Fourier transform. However, the overall effectiveness of BSARec and the roles of its individual components have yet to be systematically validated. We reproduce BSARec and show that it outperforms other SR methods on some datasets. To empirically assess whether BSARec improves performance on high-frequency signals, we propose a metric to quantify user history frequency and evaluate SR methods across different user groups. We compare digital signal processing (DSP) techniques and find that the discrete wavelet transform (DWT) offer only slight improvements over Fourier transforms, and DSP methods provide no clear advantage over simple residual connections. Finally, we explore padding strategies and find that non-constant padding significantly improves recommendation performance, whereas constant padding hinders the frequency rescaler's ability to capture high-frequency signals.