用对抗反馈环让大模型自动优化，提升可靠性。
The Meta-Prompting Protocol: Orchestrating LLMs via Adversarial Feedback Loops
- 设计生成-审计-优化三元架构，实现大模型自纠错
- 通过文本批判当梯度，减少幻觉与模型崩溃
- 适合需要高可靠性的大模型应用开发
大型语言模型从随机对话界面转向可靠软件组件，亟需重新设计交互范式。当前以启发式为主的提示工程无法提供关键应用所需的确定性保障。本文提出元提示协议，一个形式化的大模型协同控制理论框架。核心是对抗三元组：生成器（P）、审计员（A）和优化器（O）。将自然语言指令视为语义计算图中的可微变量，利用文本批评作为梯度，该架构有效缓解幻觉并防止模型坍塌。我们基于声明式编程范式（DSPy）和自动文本微分（TextGrad）验证了该方法的理论可行性，为概率计算时代的‘可观测软件工程’奠定基础。
原文摘要 · Abstract (English)
The transition of Large Language Models (LLMs) from stochastic chat interfaces to reliable software components necessitates a fundamental re-engineering of interaction paradigms. Current methodologies, predominantly heuristic-based "prompt engineering," fail to provide the deterministic guarantees required for mission-critical applications. We introduce the Meta-Prompting Protocol, a rigorous theoretical framework that formalizes the orchestration of LLMs as a programmable, self-optimizing system. Central to this protocol is the Adversarial Trinity, a tripartite topology comprising a Generator (P), an Auditor (A), and an Optimizer (O). By treating natural language instructions as differentiable variables within a semantic computation graph and utilizing textual critiques as gradients, this architecture mitigates hallucination and prevents model collapse. We demonstrate the theoretical viability of this approach using declarative programming paradigms (DSPy) and automatic textual differentiation (TextGrad), establishing a foundation for "Observable Software Engineering" in the era of probabilistic computing.
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。