EAGER-LLM: Enhancing Large Language Models as Recommenders through Exogenous Behavior-Semantic Integration

2025-02-20

AI Summary

Large language models (LLMs) suffer from inefficient collaborative learning, weak result relevance, and difficulty in fusing multi-source features in recommendation systems due to misalignment between pre-trained linguistic semantics and collaborative semantics. Method: We propose a non-intrusive behavior–semantics fusion framework that introduces (i) a dual-source knowledge-enriched item index, (ii) a multi-scale alignment reconstruction task, and (iii) an annealing adapter—enabling dual-stream (behavioral and semantic) encoding and joint contrastive-reconstructive pre-training on decoder-only LLMs without modifying the backbone. Contribution/Results: The method achieves parameter efficiency and end-to-end collaborative optimization. On three public recommendation benchmarks, it improves Recall@10 by 12.7% and NDCG@10 by 9.3% over state-of-the-art LLM-based recommenders, demonstrating the effectiveness and generalizability of semantic–collaborative co-modeling.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Sentence-level Semantics, Textual Inference, etc.
• Search and Optimization: Learning to Search

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information

Abstract

Large language models (LLMs) are increasingly leveraged as foundational backbones in the development of advanced recommender systems, offering enhanced capabilities through their extensive knowledge and reasoning. Existing llm-based recommender systems (RSs) often face challenges due to the significant differences between the linguistic semantics of pre-trained LLMs and the collaborative semantics essential for RSs. These systems use pre-trained linguistic semantics but learn collaborative semantics from scratch via the llm-Backbone. However, LLMs are not designed for recommendations, leading to inefficient collaborative learning, weak result correlations, and poor integration of traditional RS features. To address these challenges, we propose EAGER-LLM, a decoder-only llm-based generative recommendation framework that integrates endogenous and exogenous behavioral and semantic information in a non-intrusive manner. Specifically, we propose 1)dual-source knowledge-rich item indices that integrates indexing sequences for exogenous signals, enabling efficient link-wide processing; 2)non-invasive multiscale alignment reconstruction tasks guide the model toward a deeper understanding of both collaborative and semantic signals; 3)an annealing adapter designed to finely balance the model's recommendation performance with its comprehension capabilities. We demonstrate EAGER-LLM's effectiveness through rigorous testing on three public benchmarks.

Problem

• Enhance LLMs for recommender systems
• Integrate behavioral and semantic information
• Improve collaborative and semantic understanding

Innovation

• Dual-source knowledge-rich item indices
• Non-invasive multiscale alignment tasks
• Annealing adapter for performance balance