Search papers, labs, and topics across Lattice.
This paper addresses the limitations of self-attention mechanisms in sequential recommendation systems, specifically their tendency to exhibit similarity bias that hampers the capture of diverse item relations. To overcome this, the authors introduce PRISM, a module that utilizes multiple perspectives to recalibrate attention, effectively distinguishing between homogeneous and heterogeneous item relations. Extensive experiments across seven real-world datasets show that PRISM significantly enhances recommendation performance compared to existing state-of-the-art methods.
Self-attention in recommendation systems overlooks crucial item relations, but PRISM's multi-perspective approach reveals hidden user preferences that boost performance.
Capturing user preference from a user's interaction sequence is the central challenge of Sequential Recommendation (SR). This preference intuitively emerges from inter-item relations: each item transition reflects a preference embedded in the relations between items, making the faithful capture of these relations essential for accurate recommendation. For this reason, self-attention is dominant in sequential recommendation for its ability to compute pairwise item interactions, yet our empirical analysis reveals that it consistently suffers from similarity bias across various types of transformer-based SR models: dot-product attention scores disproportionately favor similar items, systematically overlooking heterogeneous relations with meaningful preference signals and directly limiting recommendation performance. To address this, we propose PRISM (Perspective-based Relational Insight Synthesis Module), a module that re-examines item relations from multiple perspectives. PRISM employs K Perspective Lenses to calibrate attention from distinct viewpoints, combining an Affinity View that refines homogeneous relations and a Contrast View that exposes heterogeneous ones suppressed by similarity bias, enabling the model to capture the full spectrum of user preferences. Extensive experiments on seven real-world benchmarks demonstrate that PRISM consistently outperforms state-of-the-art baselines. Our code is available at https://github.com/327aem/PRISM/.
