Towards Efficient Hypergraph and Multi-LLM Agent Recommender Systems
By: Tendai Mukande , Esraa Ali , Annalina Caputo and more
Potential Business Impact:
Helps online shops show you better stuff.
Recommender Systems (RSs) have become the cornerstone of various applications such as e-commerce and social media platforms. The evolution of RSs is paramount in the digital era, in which personalised user experience is tailored to the user's preferences. Large Language Models (LLMs) have sparked a new paradigm - generative retrieval and recommendation. Despite their potential, generative RS methods face issues such as hallucination, which degrades the recommendation performance, and high computational cost in practical scenarios. To address these issues, we introduce HGLMRec, a novel Multi-LLM agent-based RS that incorporates a hypergraph encoder designed to capture complex, multi-behaviour relationships between users and items. The HGLMRec model retrieves only the relevant tokens during inference, reducing computational overhead while enriching the retrieval context. Experimental results show performance improvement by HGLMRec against state-of-the-art baselines at lower computational cost.
Similar Papers
End-to-End Personalization: Unifying Recommender Systems with Large Language Models
Information Retrieval
Suggests movies you'll love, explains why.
Generative Reasoning Recommendation via LLMs
Information Retrieval
Helps computers suggest things you'll like.
Graph Retrieval-Augmented LLM for Conversational Recommendation Systems
Computation and Language
Helps computers give better suggestions by understanding you.