Leveraging Large Language Models for Sequential Recommendation

Abstract

Sequential recommendation problems have received increasing attention in research during the past few years, leading to the inception of a large variety of algorithmic approaches. In this work, we explore how large language models (LLMs), which are nowadays introducing disruptive effects in many AI-based applications, can be used to build or improve sequential recommendation approaches. Specifically, we devise and evaluate three approaches to leverage the power of LLMs in different ways. Our results from experiments on two datasets show that initializing the state-of-the-art sequential recommendation model BERT4Rec with embeddings obtained from an LLM improves NDCG by 15-20% compared to the vanilla BERT4Rec model. Furthermore, we find that a simple approach that leverages LLM embeddings for producing recommendations, can provide competitive performance by highlighting semantically related items. We publicly share the code and data of our experiments to ensure reproducibility.

arXiv comment9 pages

References in corpus (16)

• LLaMA: Open and Efficient Foundation Language Models
• Ups and Downs: Modeling the Visual Evolution of Fashion Trends with One-Class Collaborative Filtering
• Solving the apparent diversity-accuracy dilemma of recommender systems
• TALLRec: An Effective and Efficient Tuning Framework to Align Large Language Model with Recommendation
• Evaluation of Session-based Recommendation Algorithms
• BERT4Rec: Sequential Recommendation with Bidirectional Encoder Representations from Transformer
• Personalized News Recommendation with Context Trees
• Top-N Recommendation Algorithms: A Quest for the State-of-the-Art
• TabLLM: Few-shot Classification of Tabular Data with Large Language Models
• Is ChatGPT a Good Recommender? A Preliminary Study
• A Survey on Large Language Models for Recommendation
• Do LLMs Understand User Preferences? Evaluating LLMs On User Rating Prediction
• Large Language Models are Zero-Shot Rankers for Recommender Systems
• How Can Recommender Systems Benefit from Large Language Models: A Survey
• Zero-Shot Recommender Systems
• GPT4Rec: A Generative Framework for Personalized Recommendation and User Interests Interpretation

Cited by in corpus (14)

• Large Language Models for Next Point-of-Interest Recommendation
• Large Language Models for Intent-Driven Session Recommendations
• Time to Split: Exploring Data Splitting Strategies for Offline Evaluation of Sequential Recommenders
• Multi-view Intent Learning and Alignment with Large Language Models for Session-based Recommendation
• AlphaFuse: Learn ID Embeddings for Sequential Recommendation in Null Space of Language Embeddings
• Adaptive In-Context Learning with Large Language Models for Bundle Generation
• Decoy Effect In Search Interaction: Understanding User Behavior and Measuring System Vulnerability
• Uncertainty Quantification and Decomposition for LLM-based Recommendation
• LLM-RecG: A Semantic Bias-Aware Framework for Zero-Shot Sequential Recommendation
• Semantic-enhanced Co-attention Prompt Learning for Non-overlapping Cross-Domain Recommendation
• TD3: Tucker Decomposition Based Dataset Distillation Method for Sequential Recommendation
• Generative Archetype-Grounded Item Representations for Sequential Recommendation
• Empowering Denoising Sequential Recommendation with Large Language Model Embeddings
• Do We Really Need Specialization? Evaluating Generalist Text Embeddings for Zero-Shot Recommendation and Search