CoVE: Compressed Vocabulary Expansion Makes Better LLM-based Recommender Systems
This work addresses the need for more effective recommender systems in industrial applications, though it appears incremental by building on existing LLM alignment methods.
The paper tackles the problem of suboptimal performance in LLM-based recommender systems by proposing CoVE, a compressed vocabulary expansion system that assigns unique IDs to items and leverages LLMs' sequential processing, achieving significant performance enhancements on multiple datasets.
Recommender systems play a pivotal role in providing relevant content to users. With the rapid development of large language models (LLMs), researchers have begun utilizing LLMs to build more powerful recommender systems. However, existing approaches that focus on aligning LLMs with recommendation tasks do not fully leverage their sequential information processing capabilities, leading to suboptimal performance. In this paper, we propose a novel system called compressed vocabulary expansion (CoVE). In CoVE, each item is assigned a unique ID within the expanded vocabulary. Our framework effectively capitalizes on sequence understanding abilities of LLMs, significantly enhancing their performance on recommendation tasks. Additionally, we compress the embedding layer, making CoVE practical for large-scale industrial applications. The effectiveness and performance of CoVE are demonstrated through comprehensive experiments on multiple recommendation datasets and comparisons with prior works. Our code can be found at https://github.com/HaochenZhang717/CoVE-official-Repo.