Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
41 tokens/sec
GPT-4o
60 tokens/sec
Gemini 2.5 Pro Pro
44 tokens/sec
o3 Pro
8 tokens/sec
GPT-4.1 Pro
50 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

FewFedPIT: Towards Privacy-preserving and Few-shot Federated Instruction Tuning (2403.06131v2)

Published 10 Mar 2024 in cs.CR and cs.AI

Abstract: Instruction tuning has been identified as a crucial technique for optimizing the performance of LLMs in generating human-aligned responses. Nonetheless, gathering diversified and superior-quality instruction data for such tuning presents notable obstacles, especially in domains with rigid privacy provisions. Federated instruction tuning (FedIT) has emerged as a promising solution, by consolidating collaborative training across multiple data owners, thereby resulting in a privacy-preserving learning model. However, FedIT encounters limitations such as scarcity of instructional data and risk of exposure to training data extraction attacks. In this paper, we propose a novel federated algorithm, FewFedPIT, designed to simultaneously enhance privacy protection and model performance of federated few-shot learning. FewFedPITcomprises three vital components on the client side: (1) synthetic data generation, which utilizes LLMs' in-context learning capacity to generate synthetic data autonomously, thus expanding the local database; (2) parameter isolation training, which individually updates the public parameters in the synthetic data and the private parameters in the local data, consequently mitigating the noise impact of the synthetic data; (3) local aggregation sharing, which mixes public and private parameters before uploading, effectively preventing data extraction attacks. Extensive experiments on three open-source datasets demonstrate the effectiveness of FewFedPITin, enhancing privacy preservation and improving federated few-shot performance.

PDF HTML Abstract
Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize File Document Download Save Streamline Icon: https://streamlinehq.com PDF File Document Download Save Streamline Icon: https://streamlinehq.com Markdown Bookmark Chat Bubble Oval Streamline Icon: https://streamlinehq.com Chat (Pro)
Definition Search Book Streamline Icon: https://streamlinehq.com
References (41)
  1. Pythia: A suite for analyzing large language models across training and scaling. In International Conference on Machine Learning, pages 2397–2430. PMLR.
  2. Language models are realistic tabular data generators. arXiv preprint arXiv:2210.06280.
  3. What does it mean for a language model to preserve privacy? In Proceedings of the 2022 ACM Conference on Fairness, Accountability, and Transparency, pages 2280–2292.
  4. Language models are few-shot learners. Advances in neural information processing systems, 33:1877–1901.
  5. Extracting training data from large language models. In 30th USENIX Security Symposium (USENIX Security 21), pages 2633–2650.
  6. Alpagasus: Training a better alpaca with fewer data. arXiv preprint arXiv:2307.08701.
  7. Fewfedweight: Few-shot federated learning framework across multiple nlp tasks. arXiv preprint arXiv:2212.08354.
  8. Alpacafarm: A simulation framework for methods that learn from human feedback. arXiv preprint arXiv:2305.14387.
  9. Fate-llm: A industrial grade federated learning framework for large language models. arXiv preprint arXiv:2310.10049.
  10. A theoretical analysis of the repetition problem in text generation. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 35, pages 12848–12856.
  11. The pile: An 800gb dataset of diverse text for language modeling. arXiv preprint arXiv:2101.00027.
  12. Measuring the effects of non-identical data distribution for federated visual classification. arXiv preprint arXiv:1909.06335.
  13. Lora: Low-rank adaptation of large language models. arXiv preprint arXiv:2106.09685.
  14. Federated learning: Strategies for improving communication efficiency. arXiv preprint arXiv:1610.05492.
  15. Federatedscope-llm: A comprehensive package for fine-tuning large language models in federated learning. arXiv preprint arXiv:2309.00363.
  16. From quantity to quality: Boosting llm performance with self-guided data selection for instruction tuning. arXiv preprint arXiv:2308.12032.
  17. Self-alignment with instruction backtranslation. arXiv preprint arXiv:2308.06259.
  18. Chatdoctor: A medical chat model fine-tuned on a large language model meta-ai (llama) using medical domain knowledge. Cureus, 15(6).
  19. Fednlp: Benchmarking federated learning methods for natural language processing tasks. arXiv preprint arXiv:2104.08815.
  20. Chin-Yew Lin. 2004. Rouge: A package for automatic evaluation of summaries. In Text summarization branches out, pages 74–81.
  21. Communication-efficient learning of deep networks from decentralized data. In Artificial intelligence and statistics, pages 1273–1282. PMLR.
  22. Scalable extraction of training data from (production) language models. arXiv preprint arXiv:2311.17035.
  23. R OpenAI. 2023. Gpt-4 technical report. arxiv 2303.08774. View in Article, 2:13.
  24. Training language models to follow instructions with human feedback. Advances in Neural Information Processing Systems, 35:27730–27744.
  25. West-of-n: Synthetic preference generation for improved reward modeling. arXiv preprint arXiv:2401.12086.
  26. The RefinedWeb dataset for Falcon LLM: outperforming curated corpora with web data, and web data only. arXiv preprint arXiv:2306.01116.
  27. Matt Post. 2018. A call for clarity in reporting bleu scores. arXiv preprint arXiv:1804.08771.
  28. A split-and-privatize framework for large language model fine-tuning. arXiv preprint arXiv:2312.15603.
  29. Llama 2: Open foundation and fine-tuned chat models. arXiv preprint arXiv:2307.09288.
  30. Large language models are not fair evaluators. arXiv preprint arXiv:2305.17926.
  31. Self-instruct: Aligning language model with self generated instructions. arXiv preprint arXiv:2212.10560.
  32. Super-naturalinstructions: Generalization via declarative instructions on 1600+ nlp tasks. arXiv preprint arXiv:2204.07705.
  33. Openfedllm: Training large language models on decentralized private data via federated learning. arXiv preprint arXiv:2402.06954.
  34. Self-rewarding language models. arXiv preprint arXiv:2401.10020.
  35. Towards building the federated gpt: Federated instruction tuning. arXiv preprint arXiv:2305.05644.
  36. Instruction tuning for large language models: A survey. arXiv preprint arXiv:2308.10792.
  37. Alpacare: Instruction-tuned large language models for medical application. arXiv preprint arXiv:2310.14558.
  38. Fedlegal: The first real-world federated learning benchmark for legal nlp. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 3492–3507.
  39. Fedpetuning: When federated learning meets the parameter-efficient tuning methods of pre-trained language models. In Annual Meeting of the Association of Computational Linguistics 2023, pages 9963–9977. Association for Computational Linguistics (ACL).
  40. Fedprompt: Communication-efficient and privacy-preserving prompt tuning in federated learning. In ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 1–5. IEEE.
  41. Judging llm-as-a-judge with mt-bench and chatbot arena. Advances in Neural Information Processing Systems, 36.
User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (8)
  1. Zhuo Zhang (42 papers)
  2. Jingyuan Zhang (50 papers)
  3. Jintao Huang (12 papers)
  4. Lizhen Qu (68 papers)
  5. Hongzhi Zhang (33 papers)
  6. Zenglin Xu (145 papers)
  7. Qifan Wang (129 papers)
  8. Xun Zhou (62 papers)
Citations (3)
View on Semantic Scholar