Search papers, labs, and topics across Lattice.
This paper introduces DTAMLP, a novel session-based recommendation model that addresses the issue of sporadic noise in click-time intervals by integrating a weight fusion module that combines attention weights with time-interval weights. The authors demonstrate that this lightweight modification leads to consistent accuracy gains across existing time-aware and GNN-based models. Additionally, they explore the benefits of a frequency-domain filter on item embeddings, suggesting that it helps disentangle psychological preferences, thereby enhancing model performance.
A simple weight fusion technique can significantly boost session-based recommendation accuracy by filtering out noise from accidental clicks.
This paper reports two empirical findings on session-based recommendation (SBR), unified in a single model, DTAMLP. First, existing time-aware and GNN-based models (e.g., TiSASRec, SR-GNN) treat every click-time interval as equally informative, even though very short dwell times often reflect accidental clicks carrying little preference signal -- a phenomenon we call sporadic noise. We show that a lightweight, plug-and-play weight fusion module, blending a model's attention weight with a threshold-capped time-interval weight, can be inserted into such models with almost no architectural change and yields a consistent accuracy gain; we view this as the most directly verifiable contribution of this work. Second, we revisit an under-explained observation from FMLP-Rec, where a learnable frequency-domain filter on item embeddings improves accuracy, and offer a possible explanation: time-domain behavior mixes several entangled psychological preferences, and a frequency-domain view may let a model separate and down-weight such preference noise more naturally -- an interpretive conjecture rather than a proven mechanism. Building on both insights, DTAMLP, an all-MLP framework combining weight fusion and FFT-based filtering, is validated on Diginetica and RetailRocket. While this system-level design reflects the state of the field circa 2023 rather than a state-of-the-art claim, ablations confirm the two mechanisms contribute complementary, non-redundant improvements.
