用系统提示词让模型自执行流程，比复杂调度框架更高效可靠。
In-Context Prompting Obsoletes Agent Orchestration for Procedural Tasks
- 将完整流程写入提示词，让大模型自主调度执行
- 在三个任务中评分均高于4.5，错误率低于17%
- 适合追求简洁高效的多轮流程自动化应用
代理编排框架（如 LangGraph、CrewAI、Google ADK、OpenAI Agents SDK 等）在大语言模型之上设置外部调度器，用于跟踪状态并在每一步注入路由指令。我们通过受控对比发现，对于程序性任务，这种架构已被更简单的方案超越：将整个流程放入系统提示词，让模型自行调度。我们在三个领域——旅行预订（14个节点）、Zoom技术支持（14个节点）、保险理赔处理（55个节点）——各评估200次对话，采用 LLM 作为裁判，在五个质量维度上评分。基于上下文的方案得分为4.53–5.00（满分5分），而使用相同模型的 LangGraph 调度系统得分为4.17–4.84。调度系统在旅行、Zoom、保险任务中分别失败24%、9%、17%的对话，而上下文基线分别仅失败11.5%、0.5%、5%。尽管早期模型可能需要外部调度，但前沿模型能力的提升已使多轮程序化对话无需额外编排。
原文摘要 · Abstract (English)
Agent orchestration frameworks -- LangGraph, CrewAI, Google ADK, OpenAI Agents SDK, and others -- place an external orchestrator above the LLM, tracking state and injecting routing instructions at every turn. We present a controlled comparison showing that for procedural tasks, this architecture is dominated by a simpler alternative: putting the entire procedure in the system prompt and letting the model self-orchestrate. Across three domains -- travel booking (14 nodes), Zoom technical support (14 nodes), and insurance claims processing (55 nodes) -- we evaluate 200 conversations per condition using LLM-as-judge scoring on five quality criteria. The in-context approach scores 4.53--5.00 on a 5-point scale while a LangGraph orchestrator using the same model scores 4.17--4.84. The orchestrated system fails on 24% of travel, 9% of Zoom, and 17% of insurance conversations, compared to 11.5%, 0.5%, and 5% for the in-context baseline. While external orchestration may have been necessary for earlier models, advances in frontier model capabilities have made it unnecessary for multi-turn conversations following a defined procedure.
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。