Adaptive Self-Supervised Learning Strategies for Dynamic On-Device LLM Personalization
Not on the map. Why?
Abstract (arXiv)
Large language models (LLMs) have revolutionized how we interact with technology, but their personalization to individual user preferences remains a significant challenge, particularly in on-device applications. Traditional methods often depend heavily on labeled datasets and can be resource-intensive. To address these issues, we present Adaptive Self-Supervised Learning Strategies (ASLS), which utilizes self-supervised learning techniques to personalize LLMs dynamically. The framework comprises a user profiling layer for collecting interaction data and a neural adaptation layer for real-time model fine-tuning. This innovative approach enables continuous learning from user feedback, allowing the model to generate responses that align closely with user-specific contexts. The adaptive mechanisms of ASLS minimize computational demands and enhance personalization efficiency. Experimental results across various user scenarios illustrate the superior performance of ASLS in boosting user engagement and satisfaction, highlighting its potential to redefine LLMs as highly responsive and context-aware systems on-device.
Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 39 of its 50 references that are in the map. First posted on arXiv 25 Sep 2024.
Builds on
39 of 50 references.
References · 39
- 2020Language Models are Few-Shot Learnerscited by 32,199
- 2022Training language models to follow instructions with human feedbackcited by 15,712
- 2017EuroSAT: A Novel Dataset and Deep Learning Benchmark for Land Use and Land Cover Classificationcited by 1,488
- 2023Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecturecited by 982
- 2022data2vec: A General Framework for Self-supervised Learning in Speech, Vision and Languagecited by 718
- 2022Masked Autoencoders for Point Cloud Self-supervised Learningcited by 456
- 2022Scaling Vision Transformers to Gigapixel Images via Hierarchical Self-Supervised Learningcited by 357
- 2024From Persona to Personalization: A Survey on Role-Playing Language Agentscited by 206
- 2019Cross-Domain Adaptation for Animal Pose Estimationcited by 108
- 2024Can LLM be a Personalized Judge?cited by 99
- 2023Beyond Appearance: a Semantic Controllable Self-Supervised Learning Framework for Human-Centric Visual Taskscited by 94
- 2023Feature Guided Masked Autoencoder for Self-supervised Learning in Remote Sensingcited by 48
27 more
- 2024Large Language Models-guided Dynamic Adaptation for Temporal Knowledge Graph Reasoningcited by 41
- 2022On Privacy and Personalization in Cross-Silo Federated Learningcited by 36
- 2023DRDT: Dynamic Reflection with Divergent Thinking for LLM-based Sequential Recommendationcited by 27
- 2024QDyLoRA: Quantized Dynamic Low-Rank Adaptation for Efficient Large Language Model Tuningcited by 26
- 2023Enabling On-Device Large Language Model Personalization with Self-Supervised Data Selection and Synthesiscited by 20
- 2024Real-Time Pill Identification for the Visually Impaired Using Deep Learningcited by 20
- 2023Impact of Guidance and Interaction Strategies for LLM Use on Learner Performance and Perceptioncited by 19
- 2023Bongard-OpenWorld: Few-Shot Reasoning for Free-form Visual Concepts in the Real Worldcited by 18
- 2024Robust Domain Generalization for Multi-modal Object Recognitioncited by 18
- 2023Extended Agriculture-Vision: An Extension of a Large Aerial Image Dataset for Agricultural Pattern Analysiscited by 14
- 2024MobileAIBench: Benchmarking LLMs and LMMs for On-Device Use Casescited by 13
- 2023Adaptive-Solver Framework for Dynamic Strategy Selection in Large Language Model Reasoningcited by 12
- 2024Fostc3net:A Lightweight YOLOv5 Based On the Network Structure Optimizationcited by 11
- 2024Utilizing the LightGBM Algorithm for Operator User Credit Assessment Researchcited by 11
- 2022Evaluating the Fairness Impact of Differentially Private Synthetic Datacited by 9
- 2022NHA12D: A New Pavement Crack Dataset and a Comparison Study Of Crack Detection Algorithmscited by 8
- 2022On-Device Learning with Cloud-Coordinated Data Augmentation for Extreme Model Personalization in Recommender Systemscited by 6
- 2024Zero-Shot Stance Detection using Contextual Data Generation with LLMscited by 5
- 2024Root Cause Attribution of Delivery Risks via Causal Discovery with Reinforcement Learningcited by 5
- 2024What Teaches Robots to Walk, Teaches Them to Trade too -- Regime Adaptive Execution using Informed Data and LLMscited by 3
- 2024Multiple Greedy Quasi-Newton Methods for Saddle Point Problemscited by 3
- 2024Electroencephalogram Emotion Recognition via AUC Maximizationcited by 2
- 2023Multi-Task Model Personalization for Federated Supervised SVM in Heterogeneous Networkscited by 1
- 2023Large-Scale Distributed Learning via Private On-Device Locality-Sensitive Hashingcited by 1
- 2023FREEDOM: Target Label & Source Data & Domain Information-Free Multi-Source Domain Adaptation for Unsupervised Personalizationcited by 1
- 2023Parameter Efficient Tuning Allows Scalable Personalization of LLMs for Text Entry: A Case Study on Abbreviation Expansioncited by 1
- 2024Leveraging Translation For Optimal Recall: Tailoring LLM Personalization With User Profilescited by 1
Built on by
Citing papers per year: 2025: 5.
5 papers.
- 2025Adaptation of Agentic AI: A Survey of Post-Training, Memory, and Skillscited by 19
- 2025LettinGo: Explore User Profile Generation for Recommendation Systemcited by 11
- 2025NExT-Search: Rebuilding User Feedback Ecosystem for Generative AI Searchcited by 4
- 2025Advances in Large Language Models for Medicinecited by 1
- 2025When Personalization Tricks Detectors: The Feature-Inversion Trap in Machine-Generated Text Detectioncited by 1
- 2025Adaptation of Agentic AI: A Survey of Post-Training, Memory, and Skills18 Dec 2025
- 2025When Personalization Tricks Detectors: The Feature-Inversion Trap in Machine-Generated Text Detection14 Oct 2025
- 2025Advances in Large Language Models for Medicine23 Sep 2025
- 2025LettinGo: Explore User Profile Generation for Recommendation System23 Jun 2025
- 2025NExT-Search: Rebuilding User Feedback Ecosystem for Generative AI Search20 May 2025