Recommender Systems in the Era of Large Language Models (LLMs)

Abstract

With the prosperity of e-commerce and web applications, Recommender Systems (RecSys) have become an important component of our daily life, providing personalized suggestions that cater to user preferences. While Deep Neural Networks (DNNs) have made significant advancements in enhancing recommender systems by modeling user-item interactions and incorporating textual side information, DNN-based methods still face limitations, such as difficulties in understanding users' interests and capturing textual side information, inabilities in generalizing to various recommendation scenarios and reasoning on their predictions, etc. Meanwhile, the emergence of Large Language Models (LLMs), such as ChatGPT and GPT4, has revolutionized the fields of Natural Language Processing (NLP) and Artificial Intelligence (AI), due to their remarkable abilities in fundamental responsibilities of language understanding and generation, as well as impressive generalization and reasoning capabilities. As a result, recent studies have attempted to harness the power of LLMs to enhance recommender systems. Given the rapid evolution of this research direction in recommender systems, there is a pressing need for a systematic overview that summarizes existing LLM-empowered recommender systems, to provide researchers in relevant fields with an in-depth understanding. Therefore, in this paper, we conduct a comprehensive review of LLM-empowered recommender systems from various aspects including Pre-training, Fine-tuning, and Prompting. More specifically, we first introduce representative methods to harness the power of LLMs (as a feature encoder) for learning representations of users and items. Then, we review recent techniques of LLMs for enhancing recommender systems from three paradigms, namely pre-training, fine-tuning, and prompting. Finally, we comprehensively discuss future directions in this emerging field.

Accepted by IEEE TKDE

References in corpus (54)

• Survey of Hallucination in Natural Language Generation
• Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
• Graph Convolutional Neural Networks for Web-Scale Recommender Systems
• LoRA: Low-Rank Adaptation of Large Language Models
• PaLM: Scaling Language Modeling with Pathways
• A Survey of Large Language Models
• Deep Learning based Recommender System: A Survey and New Perspectives
• Self-Consistency Improves Chain of Thought Reasoning in Language Models
• LaMDA: Language Models for Dialog Applications
• TALLRec: An Effective and Efficient Tuning Framework to Align Large Language Model with Recommendation
• A Survey on the Fairness of Recommender Systems
• BloombergGPT: A Large Language Model for Finance
• A Survey on In-context Learning
• Uncovering ChatGPT's Capabilities in Recommender Systems
• Meta-Prod2Vec - Product Embeddings Using Side-Information for Recommendation
• Towards Unified Conversational Recommender Systems via Knowledge-Enhanced Prompt Learning
• Is ChatGPT Fair for Recommendation? Evaluating Fairness in Large Language Model Recommendation
• Large Language Models as Zero-Shot Conversational Recommenders
• Rethinking the Evaluation for Conversational Recommendation in the Era of Large Language Models
• DoctorGLM: Fine-tuning your Chinese Doctor is not a Herculean Task
• Chat-REC: Towards Interactive and Explainable LLMs-Augmented Recommender System
• Empowering Molecule Discovery for Molecule-Caption Translation with Large Language Models: A ChatGPT Perspective
• How to Index Item IDs for Recommendation Foundation Models
• Knowledge-enhanced Black-box Attacks for Recommendations
• Is ChatGPT a Good Recommender? A Preliminary Study
• Exploring Adapter-based Transfer Learning for Recommender Systems: Empirical Studies and Practical Insights
• Lawyer LLaMA Technical Report
• Recommender Systems with Generative Retrieval
• M6-Rec: Generative Pretrained Language Models are Open-Ended Recommender Systems
• Recommendation as Instruction Following: A Large Language Model Empowered Recommendation Approach
• When Foundation Model Meets Federated Learning: Motivations, Challenges, and Future Directions
• Do LLMs Understand User Preferences? Evaluating LLMs On User Rating Prediction
• Leveraging Large Language Models in Conversational Recommender Systems
• Beyond Chain-of-Thought, Effective Graph-of-Thought Reasoning in Language Models
• PALR: Personalization Aware LLMs for Recommendation
• A Comprehensive Survey on Trustworthy Recommender Systems
• A Brief Report on LawGPT 1.0: A Virtual Legal Assistant Based on GPT-3
• Generative Job Recommendations with Large Language Model
• Enhancing Recommender Systems with Large Language Model Reasoning Graphs
• Graph-ToolFormer: To Empower LLMs with Graph Reasoning Ability via Prompt Augmented by ChatGPT
• GPT4Rec: A Generative Framework for Personalized Recommendation and User Interests Interpretation
• Recommender AI Agent: Integrating Large Language Models for Interactive Recommendations
• Amazon-M2: A Multilingual Multi-locale Shopping Session Dataset for Recommendation and Text Generation
• Self-Generated In-Context Learning: Leveraging Auto-regressive Language Models as a Demonstration Generator
• TransRec: Learning Transferable Recommendation from Mixture-of-Modality Feedback
• Knowledge Graph Completion Models are Few-shot Learners: An Empirical Study of Relation Labeling in E-commerce with LLMs
• One4all User Representation for Recommender Systems in E-commerce
• UP5: Unbiased Foundation Model for Fairness-aware Recommendation
• A Large Language Model Enhanced Conversational Recommender System
• Automated Prompting for Non-overlapping Cross-domain Sequential Recommendation
• Large Language Models for User Interest Journeys
• Certified Robustness for Large Language Models with Self-Denoising
• LLaRA: Large Language-Recommendation Assistant
• Could Small Language Models Serve as Recommenders? Towards Data-centric Cold-start Recommendations

Cited by in corpus (15)

• When Large Language Models Meet Personalization: Perspectives of Challenges and Opportunities
• Graph Foundation Models: Concepts, Opportunities and Challenges
• A Survey on Intent-aware Recommender Systems
• EB-NeRD: A Large-Scale Dataset for News Recommendation
• CheatAgent: Attacking LLM-Empowered Recommender Systems via LLM Agent
• Decentralized LLM Inference over Edge Networks with Energy Harvesting
• The Pitfalls of Growing Group Complexity: LLMs and Social Choice-Based Aggregation for Group Recommendations
• Collaboration of Large Language Models and Small Recommendation Models for Device-Cloud Recommendation
• Advancing Tool-Augmented Large Language Models via Meta-Verification and Reflection Learning
• Just Ask for Music (JAM): Multimodal and Personalized Natural Language Music Recommendation
• Consistent Explainers or Unreliable Narrators? Understanding LLM-generated Group Recommendations
• When control meets large language models: From words to dynamics
• Effectiveness of LLMs in Temporal User Profiling for Recommendation
• FBLayout: Optimizing Memory Layout for Efficient LLM Finetuning on Mobile GPUs
• Optimizing Recall or Relevance? A Multi-Task Multi-Head Approach for Item-to-Item Retrieval in Recommendation