SEE:用于内聚上下文提示优化的策略性探索与利用
SEE: Strategic Exploration and Exploitation for Cohesive In-Context Prompt Optimization
AI总结:
针对大模型提示优化中指令与示例分离导致性能次优的问题,提出内聚上下文提示优化框架SEE,采用元启发式原理平衡探索与利用,在35个基准任务上平均性能提升13.94且计算成本降低58.67。
摘要:
为大型语言模型(LLMs)设计最优提示是一项复杂且资源密集的任务,通常需要大量的人类专业知识和精力。现有方法通常将提示指令和上下文学习示例的优化分开,导致提示缺乏内聚性,其定义和表示导致次优的任务表现。为了克服这些挑战,我们提出了一种新颖的内聚上下文提示优化框架,同时优化提示指令和示例。然而,在自然语言的离散高维空间中构建此类优化,在收敛性和计算效率方面带来了重大挑战。为了解决这些问题,我们引入了SEE,这是一个可扩展且高效的提示优化框架,采用元启发式优化原理,策略性地平衡探索与利用,以提升优化性能并实现高效收敛。SEE采用四阶段设计,在全局遍历(探索)和局部优化(利用)之间交替进行,并在优化过程中自适应选择LLM算子。我们在35个基准任务上进行了全面评估,SEE大幅优于最先进的基线方法,平均性能提升13.94,同时计算成本降低58.67。
英文摘要:
Designing optimal prompts for Large Language Models (LLMs) is a complicated and resource-intensive task, often requiring substantial human expertise and effort. Existing approaches typically separate the optimization of prompt instructions and in-context learning examples, leading to incohesive prompts that are defined and represented by suboptimal task performance. To overcome these challenges, we propose a novel Cohesive In-Context Prompt Optimization framework that refines both prompt instructions and examples. However, formulating such an optimization in the discrete and high-dimensional space of natural language poses significant challenges in both convergence and computational efficiency. To address these issues, we introduce SEE, a scalable and efficient prompt optimization framework that adopts metaheuristic optimization principles and strategically balances exploration and exploitation to enhance optimization performance and achieve efficient convergence. SEE features a quad-phased design that alternates between global traversal (exploration) and local optimization (exploitation) and adaptively chooses LLM operators during the optimization process. We have conducted a comprehensive evaluation across 35 benchmark tasks, and SEE significantly outperforms state-of-the-art baseline methods by a large margin, achieving an average performance gain of 13.94 while reducing computational costs by 58.67.
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。