Why not Collaborative Filtering in Dual View? Bridging Sparse and Dense Models

TL;DR

This paper introduces SaD, a dual-view framework combining dense and sparse models to overcome the SNR limitations of dense models in sparse data scenarios, achieving state-of-the-art recommender system performance.

Contribution

SaD is a novel, plug-and-play dual-view approach that aligns dense and sparse models to improve SNR and recommendation accuracy in sparse data environments.

Findings

SaD achieves state-of-the-art performance on benchmarks.

Dual-view alignment enhances model robustness and accuracy.

SaD outperforms strong baselines on real-world datasets.

Abstract

Collaborative Filtering (CF) remains the cornerstone of modern recommender systems, with dense embedding--based methods dominating current practice. However, these approaches suffer from a critical limitation: our theoretical analysis reveals a fundamental signal-to-noise ratio (SNR) ceiling when modeling unpopular items, where parameter-based dense models experience diminishing SNR under severe data sparsity. To overcome this bottleneck, we propose SaD (Sparse and Dense), a unified framework that integrates the semantic expressiveness of dense embeddings with the structural reliability of sparse interaction patterns. We theoretically show that aligning these dual views yields a strictly superior global SNR. Concretely, SaD introduces a lightweight bidirectional alignment mechanism: the dense view enriches the sparse view by injecting semantic correlations, while the sparse view regularizes the dense model through explicit structural signals. Extensive experiments demonstrate that, under this dual-view alignment, even a simple matrix factorization--style dense model can achieve state-of-the-art performance. Moreover, SaD is plug-and-play and can be seamlessly applied to a wide range of existing recommender models, highlighting the enduring power of collaborative filtering when leveraged from dual perspectives. Further evaluations on real-world benchmarks show that SaD consistently outperforms strong baselines, ranking first on the BarsMatch leaderboard. The code is publicly available at https://github.com/harris26-G/SaD.