Abstract
Recent advancements in large language models (LLMs) have indeed showcased their impressive capabilities. On mobile devices, the wealth of valuable, non-public data generated daily holds great promise for locally fine-tuning personalized LLMs, while maintaining privacy through on-device processing. However, the constraints of mobile device resources pose challenges to direct on-device LLM fine-tuning, mainly due to the memory-intensive nature of derivative-based optimization required for saving gradients and optimizer states. To tackle this, we propose employing derivative-free optimization techniques to enable on-device fine-tuning of LLM, even on memory-limited mobile devices. Empirical results demonstrate that the RoBERTa-large model and OPT-1.3B can be fine-tuned locally on the OPPO Reno 6 smartphone using around 4GB and 6.5GB of memory respectively, using derivative-free optimization techniques. This highlights the feasibility of on-device LLM fine-tuning on mobile devices, paving the way for personalized LLMs on resource-constrained devices while safeguarding data privacy.- Anthology ID:
- 2024.privatenlp-1.10
- Volume:
- Proceedings of the Fifth Workshop on Privacy in Natural Language Processing
- Month:
- August
- Year:
- 2024
- Address:
- Bangkok, Thailand
- Editors:
- Ivan Habernal, Sepideh Ghanavati, Abhilasha Ravichander, Vijayanta Jain, Patricia Thaine, Timour Igamberdiev, Niloofar Mireshghallah, Oluwaseyi Feyisetan
- Venues:
- PrivateNLP | WS
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 91–96
- Language:
- URL:
- https://preview.aclanthology.org/add_missing_videos/2024.privatenlp-1.10/
- DOI:
- Cite (ACL):
- Dan Peng, Zhihui Fu, and Jun Wang. 2024. PocketLLM: Enabling On-Device Fine-Tuning for Personalized LLMs. In Proceedings of the Fifth Workshop on Privacy in Natural Language Processing, pages 91–96, Bangkok, Thailand. Association for Computational Linguistics.
- Cite (Informal):
- PocketLLM: Enabling On-Device Fine-Tuning for Personalized LLMs (Peng et al., PrivateNLP 2024)
- PDF:
- https://preview.aclanthology.org/add_missing_videos/2024.privatenlp-1.10.pdf