L2Rec: Towards Dual-View Understanding of LLMs for Personalized Recommendation

2026-05-26

AI Summary

Existing large language models struggle to effectively integrate user behavioral and semantic signals for personalized recommendation, often suffering from representation misalignment due to distributional discrepancies or the absence of end-to-end supervision. This work proposes a dual-perspective personalized mixture-of-experts mechanism that unifies modeling at the parameter level for the first time. By applying low-rank perturbations to shared Transformer parameters, the method generates complementary user-specific representations from two distinct perspectives and introduces an adaptive cross-perspective fusion module to construct a unified preference representation. The entire framework supports end-to-end optimization and achieves significant performance gains over state-of-the-art baselines across four benchmark datasets, with industrial-scale A/B tests also demonstrating notable improvements in key user engagement metrics.

Abstract

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.

Problem

• personalized recommendation
• large language models
• behavioral signals
• semantic signals
• signal fusion

Innovation

• parameter-level personalization
• dual-view understanding
• low-rank adaptation
• mixture-of-experts
• LLM for recommendation