mlsanityData to 14 Oct 2026

Never Start from Scratch: Expediting On-Device LLM Personalization via Explainable Model Selection

arXiv 2504.13938 ↗First posted 15 Apr 2025cs.LG cs.AI cs.CLcited by 8

Not on the map. Why?

Abstract (arXiv)

Personalization of Large Language Models (LLMs) is important in practical applications to accommodate the individual needs of different mobile users. Due to data privacy concerns, LLM personalization often needs to be locally done at the user's mobile device, but such on-device personalization is constrained by both the limitation of on-device compute power and insufficiency of user's personal data. In this paper, we address these constraints by fine-tuning an already personalized LLM with user's personal data, and present XPerT, a new technique that ensure proper selection of such already personalized LLMs based on explainability about how they were being fine-tuned. We implemented and evaluated XPerT on various smartphone models with mainstream LLMs, and experiment results show that XPerT reduces the computation costs of on-device LLM personalization by 83%, and improves its data efficiency by 51%.

Key facts. Cited by 8 papers in the map (data through 14 Oct 2026). Builds on 51 of its 99 references that are in the map. First posted on arXiv 15 Apr 2025.

Builds on

51 of 99 references.

References · 51

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2024The Llama 3 Herd of Modelscited by 14,715
  3. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
  4. 2024Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolutioncited by 4,234
  5. 2023Generative Agents: Interactive Simulacra of Human Behaviorcited by 3,568
  6. 2022Scaling Instruction-Finetuned Language Modelscited by 3,024
  7. 2022Toy Models of Superpositioncited by 971
  8. 2017Crowdsourcing Multiple Choice Science Questionscited by 765
  9. 2016Hyperband: A Novel Bandit-Based Approach to Hyperparameter Optimizationcited by 699
  10. 2018A Discourse-Aware Attention Model for Abstractive Summarization of Long Documentscited by 478
  11. 2018BOHB: Robust and Efficient Hyperparameter Optimization at Scalecited by 470
  12. 2022DiffusionDB: A Large-scale Prompt Gallery Dataset for Text-to-Image Generative Modelscited by 379
38 more
  1. 2023Check Your Facts and Try Again: Improving Large Language Models with External Knowledge and Automated Feedbackcited by 353
  2. 2023LegalBench: A Collaboratively Built Benchmark for Measuring Legal Reasoning in Large Language Modelscited by 336
  3. 2023Emergent Linear Representations in World Models of Self-Supervised Sequence Modelscited by 319
  4. 2023Personalized Soups: Personalized Large Language Model Alignment via Post-hoc Parameter Mergingcited by 280
  5. 2020ONION: A Simple and Effective Defense Against Textual Backdoor Attackscited by 263
  6. 2024Personal LLM Agents: Insights and Survey about the Capability, Efficiency and Securitycited by 260
  7. 2023The Unlocking Spell on Base LLMs: Rethinking Alignment via In-Context Learningcited by 257
  8. 2024Continual Learning of Large Language Models: A Comprehensive Surveycited by 234
  9. 2024Two Tales of Persona in LLMs: A Survey of Role-Playing and Personalizationcited by 208
  10. 2024When Scaling Meets LLM Finetuning: The Effect of Data, Model and Finetuning Methodcited by 204
  11. 2023Evaluating Large Language Models: A Comprehensive Surveycited by 202
  12. 2023Progressive Prompts: Continual Learning for Language Modelscited by 197
  13. 2020Reformulating Unsupervised Style Transfer as Paraphrase Generationcited by 183
  14. 2020Auto-Sklearn 2.0: Hands-free AutoML via Meta-Learningcited by 174
  15. 2020Deep Learning for Text Style Transfer: A Surveycited by 162
  16. 2020MedDialog: Two Large-scale Medical Dialogue Datasetscited by 161
  17. 2022Self-Adaptive In-Context Learning: An Information Compression Perspective for In-Context Example Selection and Orderingcited by 161
  18. 2017Neural Trojanscited by 153
  19. 2023DISC-LawLLM: Fine-tuning Large Language Models for Intelligent Legal Servicescited by 120
  20. 2020MNN: A Universal and Efficient Inference Enginecited by 65
  21. 2023Goal Driven Discovery of Distributional Differences via Language Descriptionscited by 62
  22. 2023Explaining black box text modules in natural language with language modelscited by 62
  23. 2019Model selection for contextual banditscited by 60
  24. 2022Describing Differences between Text Distributions with Natural Languagecited by 54
  25. 2022FS-COCO: Towards Understanding of Freehand Sketches of Common Objects in Contextcited by 49
  26. 2023Linear Spaces of Meanings: Compositional Structures in Vision-Language Modelscited by 48
  27. 2023Bridging the Human-AI Knowledge Gap: Concept Discovery and Transfer in AlphaZerocited by 42
  28. 2017Online Learning Without Prior Informationcited by 36
  29. 2017Parameter-free online learning via model selectioncited by 34
  30. 2020Efficient Automatic CASH via Rising Banditscited by 31
  31. 2024Checkpoint Merging via Bayesian Optimization in LLM Pretrainingcited by 27
  32. 2020Online Active Model Selection for Pre-trained Classifierscited by 20
  33. 2023Real-time Neural Network Inference on Extremely Weak Devices: Agile Offloading with Explainable AIcited by 17
  34. 2023Towards Green AI in Fine-tuning Large Language Models via Adaptive Backpropagationcited by 12
  35. 2023ElasticTrainer: Speeding Up On-Device Training with Runtime Elastic Tensor Selectioncited by 12
  36. 2024MM-PhyRLHF: Reinforcement Learning Framework for Multimodal Physics Question-Answeringcited by 6
  37. 2014Mid-level Deep Pattern Miningcited by 3
  38. 2024Was it Slander? Towards Exact Inversion of Generative Language Modelscited by 3

50 most cited of 51 All 51 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 3; 2026: 5.

8 papers.

  1. 2026Agent-X: Full Pipeline Acceleration of On-device AI Agentscited by 5
  2. 2025Fints: Efficient Inference-Time Personalization for LLMs with Fine-Grained Instance-Tailored Steeringcited by 2
  3. 2025MobileFineTuner: A Mobile-Native Framework for On-Device LLM Fine-Tuning in Real-World Embedded AI Applicationscited by 2
  4. 2025PrivTune: Efficient and Privacy-Preserving Fine-Tuning of Large Language Models via Device-Cloud Collaborationcited by 1
  5. 2026Improving User Privacy in Personalized Generation: Client-Side Retrieval-Augmented Modification of Server-Side Generated Speculationscited by 1
  6. 2026TSUBASA: Improving Long-Horizon Personalization via Evolving Memory and Self-Learning with Context Distillationcited by 1
  7. 2026PowerLens: Taming LLM Agents for Safe and Personalized Mobile Power Managementcited by 0
  8. 2026FleetPilot: Fine-Grained Steering Empowers Real-World LLM Personalizationcited by 0