让基于提示的黑盒调优变得多彩：从三个正交视角提升模型泛化能力
Make Prompt-based Black-Box Tuning Colorful: Boosting Model Generalization from Three Orthogonal Perspectives
AI总结：
本文提出BBT-RGB方法，通过两阶段无导数优化、自动verbalizer构建和基于指令搜索的提示初始化三个正交组件，在少样本场景下提升黑盒调优的效率和泛化性能。
摘要：
大型语言模型（LLMs）在各类自然语言处理（NLP）任务上展现出日益强大的能力。然而，为下游任务调优这些模型通常需要高昂的成本，或因商业考虑而无法实现。近年来，黑盒调优被提出以解决该问题，其通过优化任务特定提示而无需访问梯度和隐藏表示。然而，现有大多数工作在少样本学习场景下尚未充分挖掘无梯度优化的潜力。本文提出BBT-RGB，一套直接且互补的技术，用于提升黑盒优化的效率和性能。具体而言，我们的方法包含三个即插即用组件：（1）两阶段无导数优化策略，促进快速收敛并缓解过拟合；（2）自动verbalizer构建及其在少样本设置下的新颖用法；（3）基于指令搜索和自动选择示例的更好提示初始化策略。在自然语言理解和推理的多个任务上的大量实验证明了我们方法的有效性。我们的代码已在 https://github.com/QiushiSun/BBT-RGB 公开。
英文摘要：
Large language models (LLMs) have shown increasing power on various natural language processing (NLP) tasks. However, tuning these models for downstream tasks usually needs exorbitant costs or is unavailable due to commercial considerations. Recently, black-box tuning has been proposed to address this problem by optimizing task-specific prompts without accessing the gradients and hidden representations. However, most existing works have yet fully exploited the potential of gradient-free optimization under the scenario of few-shot learning. In this paper, we describe BBT-RGB, a suite of straightforward and complementary techniques for enhancing the efficiency and performance of black-box optimization. Specifically, our method includes three plug-and-play components: (1) Two-stage derivative-free optimization strategy that facilitates fast convergence and mitigates overfitting; (2) Automatic verbalizer construction with its novel usage under few-shot settings; (3) Better prompt initialization policy based on instruction search and auto-selected demonstration. Extensive experiments across various tasks on natural language understanding and inference demonstrate the effectiveness of our method. Our codes are publicly available at https://github.com/QiushiSun/BBT-RGB.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
