This repository serves as a comprehensive survey and curated collection of research papers and resources focused on the integration of Large Language Models (LLMs) with Recommender Systems (RecSys). It aims to provide researchers and practitioners with a structured overview of how LLMs are being applied across various stages of the recommendation pipeline, from feature engineering to user interaction and system control.

How It Works

The project categorizes LLM applications in RecSys based on their role within the recommendation process: LLM for Feature Engineering (augmenting user/item features, generating instances), LLM as Feature Encoder (representation enhancement, cross-domain tasks), LLM as Scoring/Ranking Function (item scoring, item generation, hybrid tasks), LLM for User Interaction (task-oriented, open-ended), and LLM for RS Pipeline Controller. Each category lists relevant papers with details on the LLM backbone, tuning strategy, publication venue, and links.

Quick Start & Requirements

This repository is a curated list of research papers and does not involve direct code execution or installation. It requires no specific software or hardware.

Highlighted Details

• Comprehensive Taxonomy: Organizes LLM applications in RecSys into distinct functional roles within the recommendation pipeline.
• Extensive Paper Coverage: Includes a vast number of papers, categorized by their LLM application area, with links to original sources.
• Regular Updates: The "Newest Research Work List" section is updated to reflect the latest advancements in the field.
• Survey Paper: Features an accepted survey paper in ACM Transactions on Information Systems (TOIS) providing a deep dive into the topic.

Maintenance & Community

The repository is maintained by CHIANGEL/Awesome-LLM-for-RecSys. Contributions are welcomed via issues or pull requests.

Licensing & Compatibility

The repository itself is a collection of links and information, not software. Licensing is determined by the original sources of the papers.

Limitations & Caveats

This repository is a survey and does not provide executable code or implementations. Users must refer to the individual papers for practical application details.