From Overlooked to Explored: Recovering Item Relations via Mixture of Perspectives for Sequential Recommendation

• URL: http://arxiv.org/abs/2608.11846v1
• Date: Wed, 12 Aug 2026 09:39:04 GMT
• Title: From Overlooked to Explored: Recovering Item Relations via Mixture of Perspectives for Sequential Recommendation
• Abstract summary: We show that self-attention is dominant in sequential recommendation for its ability to compute pairwise item interactions.<n>We propose PRISM, a module that re-examines item relations from multiple perspectives.<n>PRISM consistently outperforms state-of-the-art baselines.
• Score: 18.117610268256005
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: 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/.

Related papers

• Adapting Knowledge Graphs for Behavior Denoising in Sequential Recommendation [9.74556536240202]
  AdaptedKG derives calibrated KG evidence for each training example without adding graph representations to the recommendation model.<n>All sample-specific scores are computed offline using training interactions and a fixed KG, so the backbone remains unchanged and no KG access is required at inference.
  arXiv  Detail & Related papers  (2026-08-21T15:54:39Z)
• RAGR: Review-Augmented Generative Recommendation [61.29879991713178]
  Review-Augmented Generative Recommendation (RAGR) is a novel GR framework that incorporates review feedback directly into the generative user sequence.<n>RAGR yields consistent and significant gains over strong GR backbones across all metrics.
  arXiv  Detail & Related papers  (2026-05-17T05:21:23Z)
• PRISM: Refracting the Entangled User Behavior Space for E-Commerce Search [5.387173334885012]
  E-commerce search systems rely on modeling user behavior to estimate item relevance and user preference.<n>User interactions are jointly shaped by exposure mechanisms, feedback loops, and semantic matching.<n>We propose PRISM, a Preference-Relevance Interaction Semantic Modeling framework for e-commerce search behavior prediction.
  arXiv  Detail & Related papers  (2026-05-08T06:05:33Z)
• User-Aware Conditional Generative Total Correlation Learning for Multi-Modal Recommendation [21.30527831237202]
  We introduce GTC, a conditional Generative Total Correlation learning framework.<n>We employ an interaction-guided diffusion model to perform user-aware content feature filtering.<n>Experiments on standard MMR benchmarks show GTC consistently outperforms state-of-the-art.
  arXiv  Detail & Related papers  (2026-04-03T12:56:58Z)
• Diversity Recommendation via Causal Deconfounding of Co-purchase Relations and Counterfactual Exposure [23.71990071700833]
  We propose Cadence: Diversity Recommendation via Causal Deconfounding of Co-purchase Relations and Counterfactual Exposure.<n>It is a plug-and-play framework built upon LightGCN to enhance recommendation diversity while preserving accuracy.<n>Our method consistently outperforms state-of-the-art diversity models in both diversity and accuracy.
  arXiv  Detail & Related papers  (2025-12-19T16:09:29Z)
• Bundle Recommendation with Item-level Causation-enhanced Multi-view Learning [1.901404011684453]
  We present BunCa, a novel bundle recommendation approach employing item-level causation-enhanced multi-view learning. BunCa provides comprehensive representations of users and bundles through two views: the Coherent View and the Cohesive View. Experiments with BunCa on three benchmark datasets demonstrate the effectiveness of this novel research.
  arXiv  Detail & Related papers  (2024-08-13T07:05:27Z)
• Invariant representation learning for sequential recommendation [0.0]
  Sequential recommendation involves automatically recommending the next item to users based on their historical item sequence. We introduce a novel sequential recommendation framework named Irl4Rec. This framework harnesses invariant learning and employs a new objective that factors in the relationship between spurious variables and adjustment variables during model training.
  arXiv  Detail & Related papers  (2023-08-22T18:39:39Z)
• When Search Meets Recommendation: Learning Disentangled Search Representation for Recommendation [56.98380787425388]
  We propose a search-Enhanced framework for the Sequential Recommendation (SESRec) SESRec disentangling similar and dissimilar representations within S&R behaviors. Experiments on both industrial and public datasets demonstrate that SESRec consistently outperforms state-of-the-art models.
  arXiv  Detail & Related papers  (2023-05-18T09:04:50Z)
• Learnable Pillar-based Re-ranking for Image-Text Retrieval [119.9979224297237]
  Image-text retrieval aims to bridge the modality gap and retrieve cross-modal content based on semantic similarities. Re-ranking, a popular post-processing practice, has revealed the superiority of capturing neighbor relations in single-modality retrieval tasks. We propose a novel learnable pillar-based re-ranking paradigm for image-text retrieval.
  arXiv  Detail & Related papers  (2023-04-25T04:33:27Z)
• GUESR: A Global Unsupervised Data-Enhancement with Bucket-Cluster Sampling for Sequential Recommendation [58.6450834556133]
  We propose graph contrastive learning to enhance item representations with complex associations from the global view. We extend the CapsNet module with the elaborately introduced target-attention mechanism to derive users' dynamic preferences. Our proposed GUESR could not only achieve significant improvements but also could be regarded as a general enhancement strategy.
  arXiv  Detail & Related papers  (2023-03-01T05:46:36Z)
• Sequential Recommendation with Auxiliary Item Relationships via Multi-Relational Transformer [74.64431400185106]
  We propose a Multi-relational Transformer capable of modeling auxiliary item relationships for Sequential Recommendation (SR) Specifically, we propose a novel self-attention module, which incorporates arbitrary item relationships and weights item relationships accordingly. Third, for inter-sequence item relationship pairs, we introduce a novel inter-sequence related items modeling module.
  arXiv  Detail & Related papers  (2022-10-24T19:49:17Z)
• Contrastive Self-supervised Sequential Recommendation with Robust Augmentation [101.25762166231904]
  Sequential Recommendationdescribes a set of techniques to model dynamic user behavior in order to predict future interactions in sequential user data. Old and new issues remain, including data-sparsity and noisy data. We propose Contrastive Self-Supervised Learning for sequential Recommendation (CoSeRec)
  arXiv  Detail & Related papers  (2021-08-14T07:15:25Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.