Home

Awesome

LLM for Recommendation Systems

A list of awesome papers and resources of recommender system on large language model (LLM).

🎉 News: Our LLM4Rec survey has been released. A Survey on Large Language Models for Recommendation

The related work and projects will be updated soon and continuously.

<div align="center"> <img src="https://github.com/WLiK/LLM4Rec-Awesome-Papers/blob/main/llm4rec_paradigms.png" alt="Editor" width="700"> </div>

If our work has been of assistance to you, please feel free to cite our survey. Thank you.

@article{llm4recsurvey,
  author       = {Likang Wu and Zhi Zheng and Zhaopeng Qiu and Hao Wang and Hongchao Gu and Tingjia Shen and Chuan Qin and Chen Zhu and Hengshu Zhu and Qi Liu and Hui Xiong and Enhong Chen},
  title        = {A Survey on Large Language Models for Recommendation},
  journal      = {CoRR},
  volume       = {abs/2305.19860},
  year         = {2023}
}

Table of Contents

The papers and related projects

No Tuning

Note: The tuning here only indicates whether the LLM model has been tuned.

NamePaperVenueYearCodeLLM
N/ALarge Language Models as Data Augmenters for Cold-Start Item RecommendationarXiv2024N/APaLM
LLM4RECLLM-Enhanced User-Item Interactions: Leveraging Edge Information for Optimized RecommendationsarXiv2024PythonGPT2
Lite-LLM4RecRethinking Large Language Model Architectures for Sequential RecommendationsarXiv2024N/AT5
Rec-GPT4VRec-GPT4V: Multimodal Recommendation with Large Vision-Language ModelsarXiv2024N/AGPT4-V, LLaVA2
LLM4VisLLM4Vis: Explainable Visualization Recommendation using ChatGPTEMNLP2023PythonGPT
LLMRecLLMRec: Large Language Models with Graph Augmentation for RecommendationWSDM2024PythonGPT
RLMRecRepresentation Learning with Large Language Models for RecommendationWWW2024PythonGPT-3.5
KP4SRKnowledge Prompt-tuning for Sequential RecommendationACM2023N/AGPT-3.5
RecInterpreterLarge Language Model Can Interpret Latent Space of Sequential RecommenderarXiv2023PythonLLaMA-7b
N/ALarge Language Models as Zero-Shot Conversational RecommendersCIKM2023PythonGPT-3.5-turbo ,GPT-4,BAIZE,Vicuna
Agent4RecOn Generative Agents in Recommendationarxiv2023PythonGPT4
N/AZero-Shot Recommendations with Pre-Trained Large Language Models for Multimodal Nudgingarxiv2023N/ABLIP-2+GPT4
InteRecAgentRecommender AI Agent: Integrating Large Language Models for Interactive Recommendationsarxiv2023N/AGPT4
GPT4SMAre GPT Embeddings Useful for Ads and Recommendation?KSEM2023PythonGPT
LLMRGEnhancing Recommender Systems with Large Language Model Reasoning Graphsarxiv2023N/AGPT-3.5/GPT4
RAHRAH! RecSys-Assistant-Human: A Human-Central Recommendation Framework with Large Language Modelsarxiv2023N/AGPT4
LLM-RecLLM-Rec: Personalized Recommendation via Prompting Large Language Modelsarxiv2023N/AGPT-3
N/ABeyond Labels: Leveraging Deep Learning and LLMs for Content MetadataRecSys2023N/AGPT4
N/ARetrieval-augmented Recommender System: Enhancing Recommender Systems with Large Language ModelsRecSys2023N/AChatGPT
N/ALLM Based Generation of Item-Description for Recommendation SystemRecSys2023N/AAlpaca
N/ALarge Language Models are Competitive Near Cold-start Recommenders for Language-and Item-based PreferencesRecSys2023N/APaLM
MINTLarge Language Model Augmented Narrative Driven RecommendationsRecsys2023N/A175B InstructGPT
KARTowards Open-World Recommendation with Knowledge Augmentation from Large Language Modelsarxiv2023PythonChatGLM
RecAgentRecAgent: A Novel Simulation Paradigm for Recommender Systemsarxiv2023PythonChatGPT
AnyPredictAnyPredict: Foundation Model for Tabular Predictionarxiv2023N/AChatGPT,BioBERT
iEvaLMRethinking the Evaluation for Conversational Recommendation in the Era of Large Language Modelsarxiv2023PythonChatGPT
N/ALarge Language Models are Zero-Shot Rankers for Recommender Systemsarxiv2023PythonChatGPT
FaiRLLMIs ChatGPT Fair for Recommendation? Evaluating Fairness in Large Language Model RecommendationRecsys2023PythonChatGPT
GENREA First Look at LLM-Powered Generative News Recommendationarxiv2023PythonChatGPT
N/ASparks of Artificial General Recommender (AGR): Early Experiments with ChatGPTarxiv2023N/AChatGPT
N/AUncovering ChatGPT's Capabilities in Recommender Systemsarxiv2023PythonChatGPT
N/AIs ChatGPT a Good Recommender? A Preliminary Studyarxiv2023N/AChatGPT
VQ-RecLearning vector-quantized item representation for transferable sequential recommendersACM2023PythonBERT
RankGPTIs ChatGPT Good at Search? Investigating Large Language Models as Re-Ranking Agentarxiv2023PythonChatGPT/4
GeneRecGenerative Recommendation: Towards Next-generation Recommender Paradigmarxiv2023PythonN/A
NIRZero-Shot Next-Item Recommendation using Large Pretrained Language Modelsarxiv2023PythonGPT-3.5
Chat-RECChat-REC: Towards Interactive and Explainable LLMs-Augmented Recommender Systemarxiv2023N/AChatGPT
N/AZero-Shot Recommendation as Language ModelingECIR2022PythonGPT-2
UniCRSTowards Unified Conversational Recommender Systems via Knowledge-Enhanced Prompt LearningKDD2022PythonGPT-2/ DialoGPT /BART
LLMRecLLMRec: Large Language Models with Graph Augmentation for RecommendationWSDM2024PythonChatGPT
K-LaMPK-LaMP: Knowledge-Guided Language Model Pre-training for Sequential RecommendationarXiv2023N/AGPT-4

Supervised Fine-Tuning

NamePaperVenueYearCodeLLM
N/AAligning Large Language Models for Controllable RecommendationsarXiv2024N/Allama2
SLIMCan Small Language Models be Good Reasoners for Sequential Recommendation?arXiv2024N/AChatGPT,llama2
GPT-FedRecFederated Recommendation via Hybrid Retrieval Augmented GenerationarXiv2024N/AE5
NoteLLMNoteLLM: A Retrievable Large Language Model for Note RecommendationWWW2024N/Allama2
N/AEnhancing Recommendation Diversity by Re-ranking with Large Language ModelsarXiv2024N/AChatGPT, LLaMA2
LLama4RecIntegrating Large Language Models into Recommendation via Mutual Augmentation and Adaptive AggregationarXiv2024N/ALLaMA
N/ALarge Language Model with Graph Convolution for RecommendationarXiv2024N/AGPT4,LLaMA2,ChatGLM
LLM-InSLarge Language Model Interaction Simulator for Cold-Start Item RecommendationarXiv2024N/ALLaMA2
LLM4RECLLM-Enhanced User-Item Interactions: Leveraging Edge Information for Optimized RecommendationsarXiv2024PythonGPT2
Fed4RecLLM-based Federated RecommendationarXiv2024N/ALLaMA
SPARSPAR: Personalized Content-Based Recommendation via Long Engagement AttentionarXiv2024N/ABERT
LLaRALLaRA: Aligning Large Language Models with Sequential Recommendersarxiv2023PythonLlama-2
E4SRecE4SRec: An Elegant Effective Efficient Extensible Solution of Large Language Models for Sequential Recommendationarxiv2023PythonLlama-2
LlamaRecLlamaRec: Two-Stage Recommendation using Large Language Models for Rankingarxiv2023PythonLlama-2
CLLM4RecCollaborative Large Language Model for Recommender Systemsarxiv2023PythonGPT2
TransRecA Multi-facet Paradigm to Bridge Large Language Model and Recommendationarxiv2023N/ABART-large and LLaMA-7B
RecMindRecMind: Large Language Model Powered Agent For RecommendationarXiv2023N/AChatGPT,P5
RecSysLLMLeveraging Large Language Models for Pre-trained Recommender Systemsarxiv2023N/AGLM-10B
N/AHeterogeneous Knowledge Fusion: A Novel Approach for Personalized Recommendation via LLMRecSys2023N/AChatGLM-6B,P5
N/APrompt Distillation for Efficient LLM-based RecommendationRecSys2023N/AT5,P5
BIGRecA Bi-Step Grounding Paradigm for Large Language Models in Recommendation Systemsarxiv2023PythonLLaMA
LLMCRSA Large Language Model Enhanced Conversational Recommender Systemarxiv2023N/AFlan-T5/LLaMA
GLRecExploring Large Language Model for Graph Data Understanding in Online Job Recommendationsarxiv2023PythonBELLE
GIRLGenerative Job Recommendations with Large Language Modelarxiv2023N/ABELLE
Amazon-M2Amazon-M2: A Multilingual Multi-locale Shopping Session Dataset for Recommendation and Text Generationarxiv2023ProjectmT5
GenRecGenRec: Large Language Model for Generative Recommendationarxiv2023PythonLLaMA
RecLLMLeveraging Large Language Models in Conversational Recommender Systemsarxiv2023N/ALaMDA(video)
ONCEONCE: Boosting Content-based Recommendation with Both Open- and Closed-source Large Language ModelsarXiv2023PythonChatGPT,Llama
DPLLMPrivacy-Preserving Recommender Systems with Synthetic Query Generation using Differentially Private Large Language Modelsarxiv2023N/AT5
PBNRPBNR: Prompt-based News Recommender Systemarxiv2023N/AT5
GPTRecGenerative Sequential Recommendation with GPTRecGen-IR@SIGIR2023N/AGPT-2
CTRLCTRL: Connect Tabular and Language Model for CTR Predictionarxiv2023N/ARoBERTa/GLM
UniTRecUniTRec: A Unified Text-to-Text Transformer and Joint Contrastive Learning Framework for Text-based RecommendationACL2023PythonBART
ICPCLarge Language Models for User Interest Journeysarxiv2023N/ALaMDA
TransRecExploring Adapter-based Transfer Learning for Recommender Systems: Empirical Studies and Practical Insightsarxiv2023N/ARoBERTa
N/AExploring the Upper Limits of Text-Based Collaborative Filtering Using Large Language Models: Discoveries and Insightsarxiv2023N/AOPT
PALRPALR: Personalization Aware LLMs for Recommendationarxiv2023N/ALLaMa
InstructRecRecommendation as instruction following: A large language model empowered recommendation approacharxiv2023N/AFLAN-T5-3B
N/ADo LLMs Understand User Preferences? Evaluating LLMs On User Rating Predictionarxiv2023N/AFLAN/ChatGPT
LSHImproving Code Example Recommendations on Informal Documentation Using BERT and Query-Aware LSH: A Comparative Studyarxiv2023N/ABERT
TALLRecTALLRec: An Effective and Efficient Tuning Framework to Align Large Language Model with Recommendationarxiv2023PythonLlama-7B
GPT4RecGPT4Rec: A Generative Framework for Personalized Recommendation and User Interests Interpretationarxiv2023N/AGPT-2
IDvs.MoRecWhere to go next for recommender systems? id-vs. modality-based recommender models revisitedSIGIR2023PythonBERT
GReaTLanguage models are realistic tabular data generatorsICLR2023PythonGPT-2
M6-RecM6-Rec: Generative Pretrained Language Models are Open-Ended Recommender Systemsarxiv2022N/AM6
N/ATowards understanding and mitigating unintended biases in language model-driven conversational recommendationInf Process Manag2023PythonBERT
P5Recommendation as language processing (rlp): A unified pretrain, personalized prompt & predict paradigm (p5)RecSys2022PythonT5
PEPLERPersonalized prompt learning for explainable recommendationTOIS2023PythonGPT-2
N/ALanguage models as recommender systems: Evaluations and limitationsNeurIPS workshop2021N/ABERT/GPT-2

Related Survey

PaperVenueYear
Exploring the Impact of Large Language Models on Recommender Systems: An Extensive Reviewarxiv2024
Large Language Models for Generative Recommendation: A Survey and Visionary Discussionsarxiv2023
Robust Recommender System: A Survey and Future Directionsarxiv2023
A Survey on Multi-Behavior Sequential Recommendationarxiv2023
When large language models meet personalization: Perspectives of challenges and opportunitiesarxiv2023
Recommender systems in the era of large language models (llms)arxiv2023
A Preliminary Study of ChatGPT on News Recommendation: Personalization, Provider Fairness, Fake Newsarxiv2023
How Can Recommender Systems Benefit from Large Language Models: A Surveyarxiv2023
Pre-train, prompt and recommendation: A comprehensive survey of language modelling paradigm adaptations in recommender systemsarxiv2023

Related Tutorial

NameVenueYear
Large Language Models for Recommendation: Progresses and Future DirectionsSIGIR-AP2023
Tutorial on Large Language Models for RecommendationRecSys2023

Common Datasets

NameSceneTasksInformationURL
Amazon ReviewCommerceSeq Rec/CF RecThis is a large crawl of product reviews from Amazon. Ratings: 82.83 million, Users: 20.98 million, Items: 9.35 million, Timespan: May 1996 - July 2014link
Amazon-M2CommerceSeq Rec/CF RecA large dataset of anonymized user sessions with their interacted products collected from multiple language sources at Amazon. It includes 3,606,249 train sessions, 361,659 test sessions, and 1,410,675 products.link
SteamGameSeq Rec/CF RecReviews represent a great opportunity to break down the satisfaction and dissatisfaction factors around games. Reviews: 7,793,069, Users: 2,567,538, Items: 15,474, Bundles: 615link
MovieLensMovieGeneralThe dataset consists of 4 sub-datasets, which describe users' ratings to movies and free-text tagging activities from MovieLens, a movie recommendation service.link
YelpCommerceGeneralThere are 6,990,280 reviews, 150,346 businesses, 200,100 pictures, 11 metropolitan areas, 908,915 tips by 1,987,897 users. Over 1.2 million business attributes like hours, parking, availability, etc.link
DoubanMovie, Music, BookSeq Rec/CF RecThis dataset includes three domains, i.e., movie, music, and book, and different kinds of raw information, i.e., ratings, reviews, item details, user profiles, tags (labels), and date.link
MINDNewsGeneralMIND contains about 160k English news articles and more than 15 million impression logs generated by 1 million users. Every news contains textual content including title, abstract, body, category, and entities.link
U-NEEDCommerceConversation RecU-NEED consists of 7,698 fine-grained annotated pre-sales dialogues, 333,879 user behaviors, and 332,148 product knowledge tuples.link
PixelRecShort VideoSeq Rec/CF RecPixelRec is a large dataset of cover images collected from a short video recommender system, comprising approximately 200 million user image interactions, 30 million users, and 400,000 video cover images. The texts and other aggregated attributes of videos are also included.link
KuaiSARVideoSearch and RecKuaiSAR contains genuine search and recommendation behaviors of 25,877 users, 6,890,707 items, 453,667 queries, and 19,664,885 actions within a span of 19 days on the Kuaishou applink
TenrecVideo, ArticleGeneralTenrec is a large-scale benchmark dataset for recommendation systems. It contains around 5 million users and 140 million interactions.link
NineRecVideo, ArticleGeneralNineRec is a TransRec dataset suite that includes a large-scale source domain recommendation dataset and nine diverse target domain recommendation datasets. Each item in NineRec is represented by a text description and a high-resolution cover image.link
MicroLensVideoGeneralMicroLens is a very large micro-video recommendation dataset containing one billion user-item interactions, 34 million users, and one million micro-videos. It includes various modality information about videos and serves as a benchmark for content-driven micro-video recommendation research.link

Single card (RTX 3090) debuggable generative language models that support Chinese corpus

Some open-source and effective projects can be adapted to the recommendation systems based on Chinese textual data. Especially for the individual researchers !

ProjectYear
Qwen-7B2023
baichuan-7B2023
YuLan-chat2023
Chinese-LLaMA-Alpaca2023
THUDM/ChatGLM-6B2023
FreedomIntelligence/LLMZoo Phoenix2023
bloomz-7b12023
LianjiaTech/BELLE2023

Hope our conclusion can help your work.

<br/>