Sparse Meets Dense: Unified Generative Recommendations With Cascaded Sparse-dense Representations

Abstract

Generative models have recently gained attention in recommendation systems by directly predicting item identifiers from user interaction sequences. However, existing methods suffer from significant information loss due to the separation of stages such as quantization and sequence modeling, hindering their ability to achieve the modeling precision and accuracy of sequential dense retrieval techniques. Integrating generative and dense retrieval methods remains a critical challenge. To address this, we introduce the Cascaded Organized Bi-Represented generAtive retrieval (COBRA) framework, which innovatively integrates sparse semantic IDs and dense vectors through a cascading process. Our method alternates between generating these representations by first generating sparse IDs, which serve as conditions to aid in the generation of dense vectors. End-to-end training enables dynamic refinement of dense representations, capturing both semantic insights and collaborative signals from user-item

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyyang2025sparse

Related papers

• Domain-adaptive And Scalable Dense Retrieval For Content-based Recommendation (2026)0.00
• Generative Retrieval As Multi-vector Dense Retrieval (2024)8.60
• Onepiece: The Great Route To Generative Recommendation -- A Case Study From Tencent Algorithm Competition (2025)0.00
• Generative Retrieval As Dense Retrieval (2023)0.00
• Hierarchical Structured Neural Network: Efficient Retrieval Scaling For Large Scale Recommendation (2024)0.00
• Cost: Contrastive Quantization Based Semantic Tokenization For Generative Recommendation (2024)7.81
• Generative Retrieval With Semantic Tree-structured Item Identifiers Via Contrastive Learning (2023)4.52
• Self-supervised Multi-modal Sequential Recommendation (2023)0.00