Dynamic Spectral Denoising with Global-Context Attention for Multi-Behavior Recommendation
2026-06-01 • Information Retrieval
Information Retrieval
AI summaryⓘ
The authors study how to recommend things better by using different types of user actions like viewing, collecting, or adding items to a cart. They find that mixing these behaviors can create confusion because some actions are noisy or unreliable for predicting what a user really wants. To fix this, they propose a method (SpectraMB) that cleans up the user behavior signals before combining them, by filtering features in a special way and using a global context to decide which signals to trust. Their experiments on real data show their method works better and is more robust to noisy behavior data.
multi-behavior recommendationrepresentation learningnoise reductionspectral filteringattention mechanismembedding spacefeature-frequency spaceuser behavior modelingrobustnessreliability calibration
Authors
Miaomiao Cai, Yunshan Ma, Fangqi Zhu, Junfeng Fang, Zhijie Zhang, Zhiyong Cheng, Xiang Wang, See-Kiong Ng
Abstract
Multi-behavior recommendation improves target-behavior prediction by exploiting heterogeneous auxiliary feedback (e.g., view, collect, and cart), yet its robustness is undermined by behavior-dependent noise and inconsistency. We argue that the key bottleneck is a representation-level failure caused by two coupled heterogeneities. First, intra-behavior representation entanglement arises when multi-hop propagation blends incidental signals with true preferences in the embedding space, making coarse spatial denoising unable to suppress noise without sacrificing informative niche signals. Second, inter-behavior reliability heterogeneity complicates cross-behavior fusion because the predictive value of auxiliary behaviors varies across users and contexts. Without reliability calibration, frequent yet unreliable signals may dominate aggregation and cause target-intent drift. To address this bottleneck, we propose Dynamic Spectral Denoising with Global-Context Attention for Multi-Behavior Recommendation (SpectraMB), a target-oriented model that performs representation purification before reliability-aware fusion. SpectraMB introduces Dynamic Feature-Level Spectral Filtering, which re-parameterizes embeddings along the feature dimension into a feature-frequency space and learns view-adaptive spectral modulation under target supervision, enabling component-wise purification without hand-crafted frequency assumptions. It further proposes Global-Context Attention Fusion, which uses a purified global representation as a context anchor to assess view compatibility and perform reliability-aware aggregation, while a residual global backbone preserves collaborative structure. Extensive experiments on three real-world datasets show that SpectraMB achieves the best results in most evaluation settings and exhibits improved robustness under noisy interactions.