# Prada: Black-Box LLM Adaptation with Private Data on Resource-Constrained Devices

arXiv: https://arxiv.org/abs/2503.14932 · HTML: https://mlsanity.com/paper/2503.14932/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 36 of its 61 references that are in the map.
- First posted on arXiv 19 Mar 2025.

## Builds on (most cited first, 30 of 36)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Communication-Efficient Learning of Deep Networks from Decentralized Data](https://mlsanity.com/paper/1602.05629/) (2016)
- [Parameter-Efficient Transfer Learning for NLP](https://mlsanity.com/paper/1902.00751/) (2019)
- [FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness](https://mlsanity.com/paper/2205.14135/) (2022)
- [Advances and Open Problems in Federated Learning](https://mlsanity.com/paper/1912.04977/) (2019)
- [Fast Inference from Transformers via Speculative Decoding](https://mlsanity.com/paper/2211.17192/) (2022)
- [Knowledge Distillation: A Survey](https://mlsanity.com/paper/2006.05525/) (2020)
- [BitFit: Simple Parameter-efficient Fine-tuning for Transformer-based Masked Language-models](https://mlsanity.com/paper/2106.10199/) (2021)
- [Towards a Unified View of Parameter-Efficient Transfer Learning](https://mlsanity.com/paper/2110.04366/) (2021)
- [AdapterFusion: Non-Destructive Task Composition for Transfer Learning](https://mlsanity.com/paper/2005.00247/) (2020)
- [Yi: Open Foundation Models by 01.AI](https://mlsanity.com/paper/2403.04652/) (2024)
- [Model Tells You What to Discard: Adaptive KV Cache Compression for LLMs](https://mlsanity.com/paper/2310.01801/) (2023)
- [Reducing Transformer Depth on Demand with Structured Dropout](https://mlsanity.com/paper/1909.11556/) (2019)
- [Parameter-Efficient Transfer Learning with Diff Pruning](https://mlsanity.com/paper/2012.07463/) (2020)
- [VeRA: Vector-based Random Matrix Adaptation](https://mlsanity.com/paper/2310.11454/) (2023)
- [Black-Box Tuning for Language-Model-as-a-Service](https://mlsanity.com/paper/2201.03514/) (2022)
- [OpenChat: Advancing Open-source Language Models with Mixed-Quality Data](https://mlsanity.com/paper/2309.11235/) (2023)
- [Integer Quantization for Deep Learning Inference: Principles and Empirical Evaluation](https://mlsanity.com/paper/2004.09602/) (2020)
- [OpenFedLLM: Training Large Language Models on Decentralized Private Data via Federated Learning](https://mlsanity.com/paper/2402.06954/) (2024)
- [FP8 Quantization: The Power of the Exponent](https://mlsanity.com/paper/2208.09225/) (2022)
- [What Matters in Transformers? Not All Attention is Needed](https://mlsanity.com/paper/2406.15786/) (2024)
- [Offsite-Tuning: Transfer Learning without Full Model](https://mlsanity.com/paper/2302.04870/) (2023)
- [LLM-CXR: Instruction-Finetuned LLM for CXR Image Understanding and Generation](https://mlsanity.com/paper/2305.11490/) (2023)
- [Extracting Prompts by Inverting LLM Outputs](https://mlsanity.com/paper/2405.15012/) (2024)
- [Counter-Interference Adapter for Multilingual Machine Translation](https://mlsanity.com/paper/2104.08154/) (2021)

## Built on by (most cited first, 2 of 2)

- [SecureInfer: Heterogeneous TEE-GPU Architecture for Privacy-Critical Tensors for Large Language Model Deployment](https://mlsanity.com/paper/2510.19979/) (2025)
- [FedMOA: Federated GRPO for Personalized Reasoning LLMs under Heterogeneous Rewards](https://mlsanity.com/paper/2602.00453/) (2026)

## Abstract (arXiv)

In recent years, Large Language Models (LLMs) have demonstrated remarkable abilities in various natural language processing tasks. However, adapting these models to specialized domains using private datasets stored on resource-constrained edge devices, such as smartphones and personal computers, remains challenging due to significant privacy concerns and limited computational resources. Existing model adaptation methods either compromise data privacy by requiring data transmission or jeopardize model privacy by exposing proprietary LLM parameters. To address these challenges, we propose Prada, a novel privacy-preserving and efficient black-box LLM adaptation system using private on-device datasets. Prada employs a lightweight proxy model fine-tuned with Low-Rank Adaptation (LoRA) locally on user devices. During inference, Prada leverages the logits offset, i.e., difference in outputs between the base and adapted proxy models, to iteratively refine outputs from a remote black-box LLM. This offset-based adaptation approach preserves both data privacy and model privacy, as there is no need to share sensitive data or proprietary model parameters. Furthermore, we incorporate speculative decoding to further speed up the inference process of Prada, making the system practically deployable on bandwidth-constrained edge devices, enabling a more practical deployment of Prada. Extensive experiments on various downstream tasks demonstrate that Prada achieves performance comparable to centralized fine-tuning methods while significantly reducing computational overhead by up to 60% and communication costs by up to 80%.