L2Rec: Towards Dual-View Understanding of LLMs for Personalized Recommendation
Not on the map. Why?
Abstract (arXiv)
Adapting large language models (LLMs) for personalized recommendation requires aligning their general-purpose capabilities with user-specific preferences while effectively leveraging both behavioral and semantic signals. Existing approaches typically integrate these signals at either the input level (e.g., injecting behavioral embeddings into the token space) or the output level (e.g., contrastive alignment of separate encoders), suffering from distribution gaps or lack of end-to-end task supervision. In this work, we introduce L2Rec, which unifies behavioral and semantic understanding at the parameter level of LLMs. Our key insight is that the same set of Transformer parameters can serve as a shared medium for both views: by applying view-specific, personalized low-rank perturbations via a Dual-view Personalized Mixture-of-Experts (DPMoE) mechanism, L2Rec enables a single LLM backbone to produce complementary behavioral and semantic adaptations for each user with minimal representation-level misalignment. An adaptive cross-view fusion module further integrates the dual-view outputs into a unified user preference. Experiments on four datasets show that L2Rec consistently outperforms state-of-the-art baselines, and online A/B testing on a large-scale industrial platform validates significant improvements in key engagement metrics.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 17 of its 18 references that are in the map. First posted on arXiv 26 May 2026.
Builds on
17 of 18 references.
References · 17
- 2025Qwen3 Technical Reportcited by 8,820
- 2017Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layercited by 3,509
- 2018Self-Attentive Sequential Recommendationcited by 1,184
- 2019BERT4Rec: Sequential Recommendation with Bidirectional Encoder Representations from Transformercited by 964
- 2024DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Modelscited by 867
- 2023Baichuan 2: Open Large-scale Language Modelscited by 670
- 2023A Survey on Large Language Models for Recommendationcited by 379
- 2020S^3-Rec: Self-Supervised Learning for Sequential Recommendation with Mutual Information Maximizationcited by 311
- 2023Adapting Large Language Models by Integrating Collaborative Semantics for Recommendationcited by 179
- 2023LLMRec: Large Language Models with Graph Augmentation for Recommendationcited by 172
- 2023Representation Learning with Large Language Models for Recommendationcited by 155
- 2023Text Is All You Need: Learning Language Representations for Sequential Recommendationcited by 149
5 more
- 2023How Can Recommender Systems Benefit from Large Language Models: A Surveycited by 142
- 2022Towards Universal Sequence Representation Learning for Recommender Systemscited by 121
- 2023Towards Open-World Recommendation with Knowledge Augmentation from Large Language Modelscited by 74
- 2023LLaRA: Large Language-Recommendation Assistantcited by 56
- 2024LEARN: Knowledge Adaptation from Large Language Model to Recommendation for Practical Industrial Applicationcited by 21
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2025EAGER-LLM: Enhancing Large Language Models as Recommenders through Exogenous Behavior-Semantic Integration
- 2024Lifelong Personalized Low-Rank Adaptation of Large Language Models for Recommendation
- 2024Direct Preference Optimization for LLM-Enhanced Recommendation Systems
- 2024Towards Efficient and Effective Unlearning of Large Language Models for Recommendation
- 2026LLM-Based User Personas for Recommendations at Scale
- 2025Improving LLM Interpretability and Performance via Guided Embedding Refinement for Sequential Recommendation
- 2024CALRec: Contrastive Alignment of Generative LLMs for Sequential Recommendation
- 2026P3Rec: Distilling Prior--Posterior Preference Reasoning for LLM-based Recommendation