mlsanityData to 14 Oct 2026

MobiEdit: Resource-efficient Knowledge Editing for Personalized On-device LLMs

arXiv 2506.13772 ↗First posted 5 Jun 2025cs.LG cs.AIcited by 1

Not on the map. Why?

Abstract (arXiv)

Large language models (LLMs) are deployed on mobile devices to power killer applications such as intelligent assistants. LLMs pre-trained on general corpora often hallucinate when handling personalized or unseen queries, leading to incorrect or outdated responses. Knowledge editing addresses this by identifying and adjusting a small crucial portion of model weights, without compromising the general knowledge. However, prior knowledge editing methods are impractical to run on local devices due to the resource-heavy backpropagation (BP) needed for updates. We present MobiEdit, the first mobile knowledge editing framework that enables efficient LLM personalization on commercial off-the-shelf (COTS) mobile devices. MobiEdit replaces full-precision BP with quantized forward-only gradient estimation, thus compatible with the energy-efficient mobile neural processing units (NPUs). MobiEdit replaces full-precision backpropagation with quantized forward-only gradient estimation, making it compatible with energy-efficient mobile NPUs. To further improve gradient estimation efficiency, we introduce two optimizations: an early stoping mechanism that adaptively terminates editing upon success and a prefix cache that reuses computation across steps. Our approach enables real-time editing of a 3B-parameter model (Qwen2.5-3B-Instruct) on COTS mobile devices with 7.6× less memory, 14.7 × less energy and 3.6× less latency compared to previous knowledge editing methods.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 16 of its 22 references that are in the map. First posted on arXiv 5 Jun 2025.

Builds on

16 of 22 references.

References · 16

  1. 2022Locating and Editing Factual Associations in GPTcited by 2,691
  2. 2022Mass-Editing Memory in a Transformercited by 1,001
  3. 2021Knowledge Neurons in Pretrained Transformerscited by 601
  4. 2021Fast Model Editing at Scalecited by 498
  5. 2024A Survey on RAG Meeting LLMs: Towards Retrieval-Augmented Large Language Modelscited by 452
  6. 2017Zero-Shot Relation Extraction via Reading Comprehensioncited by 426
  7. 2023Fine-Tuning Language Models with Just Forward Passescited by 345
  8. 2023MQuAKE: Assessing Knowledge Editing in Language Models via Multi-Hop Questionscited by 300
  9. 2024Personal LLM Agents: Insights and Survey about the Capability, Efficiency and Securitycited by 260
  10. 2013Optimal rates for zero-order convex optimization: the power of two function evaluationscited by 198
  11. 2023Knowledge Editing for Large Language Models: A Surveycited by 194
  12. 2024Small Language Models: Survey, Measurements, and Insightscited by 112
4 more
  1. 2024Model Editing Harms General Abilities of Large Language Models: Regularization to the Rescuecited by 103
  2. 2024WISE: Rethinking the Knowledge Memory for Lifelong Model Editing of Large Language Modelscited by 92
  3. 2024Fast On-device LLM Inference with NPUscited by 62
  4. 2024Lifelong Knowledge Editing for LLMs with Retrieval-Augmented Continuous Prompt Learningcited by 40

Built on by

Cited by year

Citing papers per year: 2025: 1.

1 paper.

  1. 2025SAKE: Towards Editing Auditory Attribute Knowledge of Large Audio-Language Modelscited by 5

Similar, not cited

Close in wording, no citation either way

  1. 2026EdgeFlow: Fast Cold Starts for LLMs on Mobile Devices ↗
  2. 2025MobiLLM: Enabling LLM Fine-Tuning on the Mobile Device via Server Assisted Side Tuning
  3. 2023Confidant: Customizing Transformer-based LLMs via Collaborative Edge Training
  4. 2024Learning to Edit: Aligning LLMs with Knowledge Editing
  5. 2025MobileLLM-Pro Technical Report
  6. 2026SmartRAG: Native Graph-Based RAG for Mobile Device
  7. 2024Empirical Guidelines for Deploying LLMs onto Resource-constrained Edge Devices
  8. 2026MobileMoE: Scaling On-Device Mixture of Experts