Awesome Personalized Large Language Models (PLLMs)

This repository focuses on personalized large language models (LLMs) that leverage user data to generate tailored responses based on individual user preferences. Our survey paper can be accessed on arXiv and we welcome any discussion and feedback!

• A Survey of Personalized Large Language Models: Progress and Future Directions [https://arxiv.org/abs/2502.11528]

Personalized Prompting Profile-Augmented Retrieval-Augmented Soft-Fused Contrastive Personalized Adaptation One4All One4One Personalized Alignment Data Optimization Trustworthiness Watermarking Analysis Benchmark Applications

---

Personalized Prompting

Profile-Augmented Prompting (PAG)

Method	Paper Title	Published At	Link
Cue-CoT	Cue-cot:Chain-of-thought prompting for responding to in-depth dialogue questions with llms	EMNLP'23 Findings
PAG	Integrating Summarization and Retrieval for Enhanced Personalization via Large Language Models	CIKM'23
ONCE	ONCE: Boosting Content-based Recommendation with Both Open- and Closed-source Large Language Models	WSDM'24
Matryoshka	Matryoshka: Learning to Drive Black-Box LLMs with LLMs	Arxiv'24
RewriterSlRl	Learning to Rewrite Prompts for Personalized Text Generation	WWW '24
DPL	Measuring What Makes You Unique: Difference-Aware User Modeling for Enhancing LLM Personalization	ACL'25 Findings
R2P	Reasoning Meets Personalization: Unleashing the Potential of Large Reasoning Model for Personalized Generation	Arxiv'25
EXP3RT	Review-driven Personalized Preference Reasoning with Large Language Models for Recommendation	SIGIR'25

Retrieval-Augmented Prompting (RAG)

Method	Paper Title	Published At	Link
MemPrompt	Memory-assisted prompt editing to improve GPT-3 after deployment	EMNLP'22
TeachMe	Towards Teachable Reasoning Systems: Using a Dynamic Memory of User Feedback for Continual System Improvement	EMNLP'22
MSP	Less is More: Learning to Refine Dialogue History for Personalized Dialogue Generation	ACL'22
RECAP	RECAP: Retrieval-Enhanced Context-Aware Prefix Encoder for Personalized Dialogue Response Generation	ACL'23
	Long-Term Memory for Large Language Models Through Topic-Based Vector Database	IALP'23
AuthorPred	Teach LLMs to Personalize -- An Approach inspired by Writing Education	Arxiv'23
MaLP	LLM-based Medical Assistant Personalization with Short- and Long-Term Memory Coordination	NAACL'24
MemoRAG	MemoRAG: Moving towards Next-Gen RAG Via Memory-Inspired Knowledge Discovery	WebConf'25
PEARL	Pearl: Personalizing Large Language Model Writing Assistants with Generation-Calibrated Retrievers	CustomNLP4U@ EMNLP'24
CoPS	Cognitive Personalized Search Integrating Large Language Models with an Efficient Memory Mechanism	WebConf'24
LaMP	LaMP: When Large Language Models Meet Personalization	ACL'24
ROPG / RSPG	Optimization Methods for Personalizing Large Language Models through Retrieval Augmentation	SIGIR'24
HYDRA	HYDRA: Model Factorization Framework for Black-Box LLM Personalization	NeurIPS'24
TACITREE	Toward Multi-Session Personalized Conversation: A Large-Scale Dataset and Hierarchical Tree Framework for Implicit Reasoning	Arxiv'25
Persona-DB	Persona-DB: Efficient Large Language Model Personalization for Response Prediction with Collaborative Data Refinement	COLING'25
CFRAG	Retrieval Augmented Generation with Collaborative Filtering for Personalized Text Generation	SIGIR'25
AP-Bots	Improving RAG for Personalization with Author Features and Contrastive Examples	Arxiv'25
FERMI	Few-shot Personalization of LLMs with Mis-aligned Responses	NAACL'25
RPM	(LLMs Think, But Not In Your Flow: Reasoning-Level Personalization for Black-Box Large Language Models.) RPM: Reasoning-Level Personalization for Black-Box Large Language Models	Arxiv'25
PRIME	Large Language Model Personalization with Cognitive Memory and Thought Processes	Arxiv'25
PersonaAgent	PersonaAgent: When Large Language Model Agents Meet Personalization at Test Time	Arxiv'25
Mem0	Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory	Arxiv'25
MemWeaver	MemWeaver: A Hierarchical Memory from Textual Interactive Behaviors for Personalized Generation	Arxiv'25

Soft-Fused Prompting

Method	Paper Title	Published At	Link
GSMN	Personalized Response Generation via Generative Split Memory Network	NAACL'21
RECAP	RECAP: Retrieval-Enhanced Context-Aware Prefix Encoder for Personalized Dialogue Response Generation	ACL'23
UEM	User Embedding Model for Personalized Language Prompting	PERSONALIZE@ ACL'24
PERSOMA	PERSOMA: PERsonalized SOft ProMpt Adapter Architecture for Personalized Language Prompting	GenAIRecP@ KDD'24
REGEN	Beyond Retrieval: Generating Narratives in Conversational Recommender Systems	Arxiv'24
PeaPOD	Preference Distillation for Personalized Generative Recommendation	Arxiv'24
User-LLM	User-LLM: Efficient LLM Contextualization with User Embeddings	Arxiv'24
ComMer	ComMer: a Framework for Compressing and Merging User Data for Personalization	Arxiv'25
PPlug	LLMs + Persona-Plug = Personalized LLMs	ACL'25
DEP	Latent Inter-User Difference Modeling for LLM Personalization	EMNLP'25
CBP-Tuning	CBP-Tuning: Efficient Local Customization for Black-box Large Language Models	Arxiv'25
UniCTokens	UniCTokens: Boosting Personalized Understanding and Generation via Unified Concept Tokens	NeurIPS'25

Contrastive Prompting

Method	Paper Title	Published At	Link
CoS	(CoS: Enhancing Personalization and Mitigating Bias with Context Steering) Context Steering: Controllable Personalization at Inference Time	(arXiv'24) ICLR'25
StyleVector	Personalized Text Generation with Contrastive Activation Steering	ACL'25
CoSteer	CoSteer: Collaborative Decoding-Time Personalization via Local Delta Steering	Arxiv'25
Fints	Fints: Efficient Inference-Time Personalization for LLMs with Fine-Grained Instance-Tailored Steering	Arxiv'25
SteerX	SteerX: Disentangled Steering for LLM Personalization	Arxiv'25

Personalized Adaptation

One PEFT All Users (One4All)

Method	Paper Title	Published At	Link
UserIdentifier	UserIdentifier: Implicit User Representations for Simple and Effective Personalized Sentiment Analysis	FL4NLP @ ACL'22
PEFT-U	PEFT-U: Parameter-Efficient Fine-Tuning for User Personalization	Arxiv'24
PLoRA	Personalized LoRA for Human-Centered Text Understanding	AAAI'24
LM-P / CLS-P	Personalized Large Language Models	SENTIRE@ ICDM'24
Review-LLM	Review-LLM: Harnessing Large Language Models for Personalized Review Generation	Arxiv'24
MiLP	Personalized LLM Response Generation with Parameterized Memory Injection	Arxiv'24
IGD	IGD: Token Decisiveness Modeling via Information Gain in LLMs for Personalized Recommendation	NeurIPS'25

One PEFT Per Users (One4One)

Method	Paper Title	Published At	Link
UserAdapter	UserAdapter: Few-Shot User Learning in Sentiment Analysis	ACL'21 Findings
RecLoRA	Lifelong Personalized Low-Rank Adaptation of Large Language Models for Recommendation	Arxiv'24
iLoRA	Customizing Language Models with Instance-wise Lora for Sequential Recommendation	NeurIPS'24
PocketLLM	PocketLLM: Enabling On-Device Fine-Tuning for Personalized LLMs	PrivateNLP@ ACL'24
OPPU	Democratizing Large Language Models via Personalized Parameter-Efficient Fine-tuning	EMNLP'24
Per-Pcs	Personalized Pieces: Efficient Personalized Large Language Models through Collaborative Efforts	EMNLP'24
	Personalized Collaborative Fine-Tuning for LLMs	COLM'24
FDLoRA	FDLoRA: Personalized Federated Learning of Large Language Model via Dual LoRA Tuning	Arxiv'24
HYDRA	HYDRA: Model Factorization Framework for Black-Box LLM Personalization	NeurIPS'24
PROPER	PROPER: A Progressive Learning Framework for Personalized Large Language Models with Group-Level Adaptation	ACL'25
HomeLLaMA	Towards Privacy-Preserving and Personalized Smart Homes via Tailored Small Language Models	(Arxiv'25) IEEE TMC'26
MobiEdit	MobiEdit: Resource-efficient Knowledge Editing for Personalized On-device LLMs	Arxiv'25
CM	Unlearning Misalignment for Personalized LLM Adaptation via Instance-Response-Dependent Discrepancies	TMLR(09/2025), Journal Submissions
Profile-to-PEFT	Instant Personalized Large Language Model Adaptation via Hypernetwork	Arxiv'25
CoMiGS	On-Device Collaborative Language Modeling via a Mixture of Generalists and Specialists	ICML'25

Personalized Alignment

Data Construction

Method	Paper Title	Published At	Link
PLUM	On the Way to LLM Personalization: Learning to Remember User Conversations	(Arxiv'24) ACL Workshop'25
	Aligning to Thousands of Preferences via System Message Generalization	NeurIPS'24
	Enabling On-Device Large Language Model Personalization with Self-Supervised Data Selection and Synthesis	DAC'24
PRISM	The PRISM Alignment Dataset: What Participatory, Representative and Individualised Human Feedback Reveals About the Subjective and Multicultural Alignment of Large Language Models	NeurIPS'24 (Best Paper)
PersonalLLM	PersonalLLM: Tailoring LLMs to Individual Preferences	Arxiv'24
	Aligning LLMs with Individual Preferences via Interaction	COLING'25

Optimization

Method	Paper Title	Published At	Link
MORLHF	Fine-Grained Human Feedback Gives Better Rewards for Language Model Training	NeurIPS'23
Personalized Soups	Personalized Soups: Personalized Large Language Model Alignment via Post-hoc Parameter Merging	Arxiv'23, NeurIPS Workshop'23
Reward Soups	Rewarded soups: towards Pareto-optimal alignment by interpolating weights fine-tuned on diverse rewards	NuerIPS'23
MODPO	Beyond One-Preference-Fits-All Alignment: Multi-Objective Direct Preference Optimization	ACL'24 Findings
MOD	Decoding-Time Language Model Alignment with Multiple Objectives	NeurIPS'24
BiPO	Personalized Steering of Large Language Models: Versatile Steering Vectors Through Bi-directional Preference Optimization	NeurIPS'24
PPT	Personalized Adaptation via In-Context Preference Learning	Arxiv'24, NeurIPS Workshop'24
VPL	Personalizing Reinforcement Learning from Human Feedback with Variational Preference Learning	NeurIPS'24
CIPHER, PRELUDE	Aligning LLM Agents by Learning Latent Preference from User Edits	NeurIPS'24
CHAMELEON	Personalize Your LLM: Fake it then Align it	NAACL'25 Findings
REST-PG	Reasoning-Enhanced Self-Training for Long-Form Personalized Text Generation	Arxiv'25
Drift	Drift: Decoding-time Personalized Alignments with Implicit User Preferences	Arxiv'25, EMNLP Findings'25
PAD	PAD: Personalized Alignment of LLMs at Decoding-Time	ICLR'25
Amulet	Amulet: ReAlignment During Test Time for Personalized Preference Adaptation of LLMs	ICLR'25
RLPA	Teaching Language Models to Evolve with Users: Dynamic Profile Modeling for Personalized Alignment	Arxiv'25
PROSE	Aligning LLMs by Predicting Preferences from User Writing Samples	ICML'25
DEEPER	DEEPER Insight into Your User: Directed Persona Refinement for Dynamic Persona Modeling	ACL'25
NextQuill	NextQuill: Causal Preference Modeling for Enhancing LLM Personalization	ICLR'26 (Arxiv'25)
RePIC	RePIC: Reinforced Post-Training for Personalizing Multi-Modal Language Models	NeurIPS'25
UserAlign	Inference-Time Personalized Alignment with a Few User Preference Queries	NeurIPS'25
CURIO	Enhancing Personalized Multi-Turn Dialogue with Curiosity Reward	NeurIPS'25

System (Agent)

Method	Paper Title	Published At	Link
LD_Agent	Hello Again! LLM-powered Personalized Agent for Long-term Dialogue	NAACL'25
PUMA, PersonalWAB Benchmark	Large Language Models Empowered Personalized Web Agents	WWW'25
EgoMem	EgoMem: Lifelong Memory Agent for Full-duplex Omnimodal Models	Arxiv'25

Trustworthiness

Watermarking

Method	Paper Title	Published At	Link
SAEMark	SAEMark: Shaping Signatures with Adaptable Encoder for Robust and Imperceptible Watermarking	NeurIPS'25

Analysis

Keyword	Paper Title	Published At	Link
Role of User Profile	Understanding the Role of User Profile in the Personalization of Large Language Models	Arxiv'24
RAG vs. PEFT	Comparing Retrieval-Augmentation and Parameter-Efficient Fine-Tuning for Privacy-Preserving Personalization of Large Language Models	(Arxiv'24) ICTIR'25
Safety-Utility	Exploring Safety-Utility Trade-Offs in Personalized Language Models	(Arxiv'24) NAACL'24
PerCRS	Exploring the Impact of Personality Traits on Conversational Recommender Systems: A Simulation with Large Language Models	Arxiv'25

Metric

Name	Paper Title	Published At	Link
EGISES	Accuracy is Not Enough: Evaluating Personalization in Summarizers	EMNLP'23 Findings
PerSEval	PerSEval: Assessing Personalization in Text Summarizers	TMLR'24
ExPerT	ExPerT: Effective and Explainable Evaluation of Personalized Long-Form Text Generation	ACL'25 Findings

Benchmark

Name	Paper Title	Published At	Link
PER-CHAT	Personalized Response Generation via Generative Split Memory Network	NAACL'21
LaMP	LaMP: When Large Language Models Meet Personalization	ACL'24
LongLaMP	LongLaMP: A Benchmark for Personalized Long-form Text Generation	Arxiv'24
PerLTQA	PerLTQA: A Personal Long-Term Memory Dataset for Memory Classification, Retrieval, and Fusion in Question Answering	SIGHAN@ ACL'24
PEFT-U	PEFT-U: Parameter-Efficient Fine-Tuning for User Personalization	Arxiv'24
REGEN	Beyond Retrieval: Generating Narratives in Conversational Recommender Systems	Arxiv'24
PRISM	The PRISM Alignment Dataset: What Participatory, Representative and Individualised Human Feedback Reveals About the Subjective and Multicultural Alignment of Large Language Models	NeurIPS'24 Best Paper Award
PersonalLLM	PersonalLLM: Tailoring LLMs to Individual Preferences	ICLR'25
ALOE	Aligning LLMs with Individual Preferences via Interaction	COLING'25
PGraphRAG	Personalized Graph-Based Retrieval for Large Language Models	Arxiv'25
PrefEval	Do LLMs Recognize Your Preferences? Evaluating Personalized Preference Following in LLMs	ICLR'25
LongMemEval	LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory	ICLR'25
PersonalWAB	Large Language Models Empowered Personalized Web Agents	WWW'25
IMPLEXCONV	Toward Multi-Session Personalized Conversation: A Large-Scale Dataset and Hierarchical Tree Framework for Implicit Reasoning	Arxiv'25
PERSONACONVBENCH	A Personalized Conversational Benchmark: Towards Simulating Personalized Conversations	Arxiv'25
HiCUPID	Exploring the Potential of LLMs as Personalized Assistants: Dataset, Evaluation, and Analysis	ACL'25
LaMP-QA	LaMP-QA: A Benchmark for Personalized Long-form Question Answering	Arxiv'25
DPL	Measuring What Makes You Unique: Difference-Aware User Modeling for Enhancing LLM Personalization	ACL'25 Findings
REALTALK	REALTALK: A 21-Day Real-World Dataset for Long-Term Conversation	Arxiv'25
MemBench	MemBench: Towards More Comprehensive Evaluation on the Memory of LLM-based Agents	ACL'25 Findings
PersonaMem	Know Me, Respond to Me: Benchmarking LLMs for Dynamic User Profiling and Personalized Responses at Scale	(Arxiv'25) COLM'25
MemoryAgentBench	Evaluating Memory in LLM Agents via Incremental Multi-Turn Interactions	(Arxiv'25) ICLR'26
PersonaLens	PersonaLens: A Benchmark for Personalization Evaluation in Conversational AI Assistants	(Arxiv'25) ACL Findings'25
Theanine	Towards Lifelong Dialogue Agents via Timeline-based Memory Management	NAACL'25
PrefDisco	Personalized Reasoning: Just-in-Time Personalization and Why LLMs Fail at It	(Arxiv'25) ICLR'26
MMPB	MMPB: It’s Time for Multi-Modal Personalization	NeurIPS'25
UnifyBench	UniCTokens: Boosting Personalized Understanding and Generation via Unified Concept Tokens (introduces UnifyBench)	NeurIPS'25
AgentRecBench	AgentRecBench: Benchmarking LLM Agent-based Personalized Recommender Systems	NeurIPS'25
BenchPreS	BenchPreS: A Benchmark for Context-Aware Personalized Preference Selectivity of Persistent-Memory LLMs	arXiv'26

Applications

Name	Paper Title	Published At	Link	Domain
HumSum	HumSum: A Personalized Lecture Summarization Tool for Humanities Students Using LLMs	ACL Workshop'24		Humanity
LearnLens	LearnLens: LLM-Enabled Personalised, Curriculum-Grounded Feedback with Educators in the Loop	(Arxiv'25) EMNLP Demo'25		Education
Divide and Conquer	Adapting and Evaluating Multimodal Large Language Models for Adolescent Idiopathic Scoliosis Self-Management: A Divide and Conquer Framework	(Arxiv'25) AI for Clinical Applications'25		Healthcare
AdImpact-RL	Learning Personalized Ad Impact via Contextual Reinforcement Learning under Delayed Rewards	NeurIPS'25		Advertising
FreeCure	Foundation Cures Personalization: Improving Personalized Models’ Prompt Consistency via Hidden Foundation Knowledge	NeurIPS'25		Diffusion Personalization

Related Workshops

Workshop Name	Conference/Series	Link
The First Workshop on Personalized Generative AI @ CIKM 2023: Personalization Meets Large Language Models.	CIKM'23
The 1st Workshop on Personalization of Generative AI Systems	ACL'24
The 3rd Workshop on Personal Intelligence with Generative AI	WWW'25
LLM4Good: Sustainable & Trustworthy LLMs for Personalization	ACM UMAP'25
P13N: Personalization in Generative AI Workshop	ICCV'25
PerFM: Personalization in the Era of Large Foundation Models	AAAI'26

Citation

@article{liu2025survey,
  title={A Survey of Personalized Large Language Models: Progress and Future Directions},
  author={Liu, Jiahong and Qiu, Zexuan and Li, Zhongyang and Dai, Quanyu and Zhu, Jieming and Hu, Minda and Yang, Menglin and King, Irwin},
  journal={arXiv preprint arXiv:2502.11528},
  year={2025}
}