GenRec: Large Language Model for Generative Recommendation

Jianchao Ji, Zelong Li, Shuyuan Xu, Wenyue Hua, Yingqiang Ge, Juntao Tan, Yongfeng Zhang

Abstract: In recent years, Large Language Models (LLMs) have emerged as powerful tools for diverse natural language processing tasks. However, their potential for recommender systems under the generative recommendation paradigm remains relatively unexplored. This paper presents an innovative approach to recommender systems using Large Language Models (LLMs) purely based on raw text data, i.e., using item name or title as item IDs rather than creating meticulously designed user or item IDs. More specifically, we present a novel LLM for Generative Recommendation (GenRec) method that utilizes the expressive power of LLM to directly generate the target item to recommend, rather than calculating the ranking score for each candidate item one by one as in traditional discriminative recommendation. GenRec uses LLM’s understanding ability to interpret context, learn user preferences, and generate relevant recommendations. Our proposed approach leverages the vast knowledge encoded in Large Language Models to accomplish recommendation tasks. We formulate specialized prompts to enhance the ability of LLM to comprehend recommendation tasks. Subsequently, we use these prompts to LoRA-fine-tune the LLaMA backbone LLM on the user-item interaction data represented by raw text (using raw item name or title as the item’s ID) to capture user preferences and item characteristics. Our research underscores the potential of LLM-based generative recommendation in revolutionizing the domain of recommendation systems and offers a foundational framework for future explorations in this field. We conduct extensive experiments on benchmark datasets, and the experiments show that our GenRec method achieves better results on large datasets. Code and data are are open-source at GitHub (https://github.com/rutgerswiselab/GenRec).