RAT 论文提出可动态重组的智能体框架形式化基础
Reconfigurable Action Theories: A Formal Foundation for Dynamically Composable Agent Harnesses
做智能体框架的开发者可以看看这篇。它把 OpenHands、DeepSeek Harness 这类运行中换工具、换策略的行为用形式化语言描述清楚，还能编译成 ASP 验证。如果你的 harness 会动态挂载 MCP 工具或切换模型，这套理论能帮你检查组件变更后行为是否一致。
一篇 arXiv 论文（编号 2611.12300）提出 Reconfigurable Action Theories（RAT），用于形式化描述智能体 harness 在运行时更换工具、提供商和策略的行为。传统 action theory 固定词汇表和因果律，而 RAT 允许动作本身修改理论：每个模块包含词汇表、因果律和初始化器，模块的加载与卸载由普通因果律表示的动作完成。
论文给出三条充分条件：下一配置唯一、新增模块不破坏既有行为、模块以不同顺序加载结果相同；同时证明了 RAT 推理任务的计算复杂度，并将有限 RAT 系统翻译为非单调因果理论和 Answer Set Programming（ASP），保留有界执行。实验将有限 RAT 模型与 DeepSeek Harness、Pi 和 OpenHands 官方实现的执行过程进行对比。
这项工作为检查组件变更如何影响 harness 的可用动作、行为和可达目标提供了形式化基础，填补了动态智能体框架缺乏严格语义定义的空白。
Reconfigurable Action Theories: A Formal Foundation for Dynamically Composable Agent Harnesses
Action theories typically fix their vocabulary and causal laws, whereas agent harnesses can change their tools, providers, and policies during execution. We introduce Reconfigurable Action Theories (RAT), in which actions can change the action theory itself. A theory module in RAT contains a vocabulary, causal laws, and an initializer. A RAT configuration records the active modules and the world state. Module loading and unloading are represented by actions governed by ordinary causal laws. RAT retains the standard causal action semantics of its language fragment when the active module set is fixed. We give sufficient conditions for at most one next configuration, preservation of existing behavior when a module is added, and equal results when modules are loaded in either order. We establish the computational complexity of reasoning tasks in RAT and translate finite RAT systems into nonmonotonic causal theories and Answer Set Programming (ASP), preserving bounded executions. Experiments compare finite RAT models with executions of official DeepSeek Harness, Pi, and OpenHands implementations. RAT provides a formal basis for checking how component changes affect a harness's available actions, behavior, and reachable goals.
本条情报归属 AITOP「论文」核心栏目，重点聚焦北美前沿实验室与硅谷科技巨头。在跨区域产业观察坐标中，该动态直接关联「AI 开发者工具链、Coding Agent 与工程化实践」研究主线，反映出当前技术演进与商业落地的最新风向。涉及核心实体与技术议题包括：RAT、智能体、Answer Set Programming、OpenHands，读者可通过下方标签进入对应专题页查阅历史脉络与关联对比。
信源核验说明：本文原始信息发布自一手权威信源「arXiv: DeepSeek」（Tier-1 一手发布），具备高确定性参考价值。