无需访问模型权重或上传数据，本地高效定制大模型。
CBP-Tuning: Efficient Local Customization for Black-box Large Language Models
- 服务器端训练提示生成器，用户端无梯度优化软提示。
- 每任务仅需一个定制向量，实现高效个性化适配。
- 保护隐私且在多个领域表现优于基线方法。
大语言模型的定制成本高昂，限制了其适应用户需求的能力。为此，我们提出定制黑箱提示调优（CBP-Tuning），一种高效本地定制框架，同时保障双向隐私。该框架分两阶段：(1) 服务端训练提示生成器，捕获领域特异性与任务无关能力；(2) 用户端无梯度优化，为个体任务微调软提示。该方法避免用户访问模型权重或上传私密数据，仅需每个任务一个定制向量即可实现有效适配。在常识推理、医疗和金融领域的评估表明，相比基线方法，CBP-Tuning 在任务无关处理与隐私保护方面均具优势。
原文摘要 · Abstract (English)
The high costs of customizing large language models (LLMs) fundamentally limit their adaptability to user-specific needs. Consequently, LLMs are increasingly offered as cloud-based services, a paradigm that introduces critical limitations: providers struggle to support personalized customization at scale, while users face privacy risks when exposing sensitive data. To address this dual challenge, we propose Customized Black-box Prompt Tuning (CBP-Tuning), a novel framework that facilitates efficient local customization while preserving bidirectional privacy. Specifically, we design a two-stage framework: (1) a prompt generator trained on the server-side to capture domain-specific and task-agnostic capabilities, and (2) user-side gradient-free optimization that tailors soft prompts for individual tasks. This approach eliminates the need for users to access model weights or upload private data, requiring only a single customized vector per task while achieving effective adaptation. Furthermore, the evaluation of CBP-Tuning in the commonsense reasoning, medical and financial domain settings demonstrates superior performance compared to baselines, showcasing its advantages in task-agnostic processing and privacy preservation.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。