SEMINAR: Search Enhanced Multi-modal Interest Network And Approximate Retrieval For Lifelong Sequential Recommendation

Abstract

The modeling of users' behaviors is crucial in modern recommendation systems. A lot of research focuses on modeling users' lifelong sequences, which can be extremely long and sometimes exceed thousands of items. These models use the target item to search for the most relevant items from the historical sequence. However, training lifelong sequences in click through rate (CTR) prediction or personalized search ranking (PSR) is extremely difficult due to the insufficient learning problem of ID embedding, especially when the IDs in the lifelong sequence features do not exist in the samples of training dataset. Additionally, existing target attention mechanisms struggle to learn the multi-modal representations of items in the sequence well. The distribution of multi-modal embedding (text, image and attributes) output of user's interacted items are not properly aligned and there exist divergence across modalities. We also observe that users' search query sequences and item browsing sequences

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyshen2024seminar

Related papers

• Self-supervised Multi-modal Sequential Recommendation (2023)0.00
• R2LED: Equipping Retrieval And Refinement In Lifelong User Modeling With Semantic Ids For CTR Prediction (2026)0.00
• Raserec: Retrieval-augmented Sequential Recommendation (2024)2.92
• Optimizing Recall Or Relevance? A Multi-task Multi-head Approach For Item-to-item Retrieval In Recommendation (2025)0.00
• Joint Modeling And Optimization Of Search And Recommendation (2018)0.00
• Everyone's Preference Changes Differently: Weighted Multi-interest Retrieval Model (2022)0.00
• Benefit From Rich: Tackling Search Interaction Sparsity In Search Enhanced Recommendation (2025)0.00
• Domain-adaptive And Scalable Dense Retrieval For Content-based Recommendation (2026)0.00