ZERA：零初始化指令进化精炼智能体——通过基于原则的优化从零指令到结构化提示
ZERA: Zero-init Instruction Evolving Refinement Agent -- From Zero Instructions to Structured Prompts via Principle-based Optimization
AI总结：
ZERA提出一种联合优化系统与用户提示的框架，通过八个可泛化标准自动加权评分并结构化修订，实现小样本、短周期内快速收敛，在五个LLM和九个数据集上超越强基线。
摘要：
自动提示优化（APO）通过为特定任务精炼提示来提升大语言模型（LLM）的性能。然而，先前的APO方法通常仅关注用户提示，依赖非结构化的反馈，并且需要大样本量和长迭代周期——这使得它们成本高昂且脆弱。我们提出ZERA（零初始化指令进化精炼智能体），一种新颖的框架，通过基于原则的低开销精炼来联合优化系统提示和用户提示。ZERA使用八个可泛化的标准对提示进行评分，这些标准具有自动推断的权重，并根据这些结构化批评来修订提示。这使得在最小样本量和短迭代周期内快速收敛到高质量提示成为可能。我们在五个LLM和九个涵盖推理、摘要和代码生成任务的不同数据集上评估了ZERA。实验结果表明，相对于强基线有持续改进。进一步的消融研究突出了每个组件对更有效提示构建的贡献。我们的实现包括所有提示，公开于https://github.com/younatics/zera-agent。
英文摘要：
Automatic Prompt Optimization (APO) improves large language model (LLM) performance by refining prompts for specific tasks. However, prior APO methods typically focus only on user prompts, rely on unstructured feedback, and require large sample sizes and long iteration cycles-making them costly and brittle. We propose ZERA (Zero-init Instruction Evolving Refinement Agent), a novel framework that jointly optimizes both system and user prompts through principled, low-overhead refinement. ZERA scores prompts using eight generalizable criteria with automatically inferred weights, and revises prompts based on these structured critiques. This enables fast convergence to high-quality prompts using minimal examples and short iteration cycles. We evaluate ZERA across five LLMs and nine diverse datasets spanning reasoning, summarization, and code generation tasks. Experimental results demonstrate consistent improvements over strong baselines. Further ablation studies highlight the contribution of each component to more effective prompt construction. Our implementation including all prompts is publicly available at https://github.com/younatics/zera-agent.
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。
