原帖引用 · AI 翻译
标题：“大型语言模型智能体之间的零样本协调”
牛津大学的一篇新论文发现，前沿大型语言模型智能体无法与不熟悉的伙伴协调，因为每个智能体都把自己臆定的规则（任意约定）当作显而易见的选择。
不同公司的智能体越来越需要在没有商定规则的情况下合作。以往研究使用共享的运行框架或复杂游戏，掩盖了协调失败的原因。
他们在 9 个小型合作游戏中测试了基于 GPT-5.2 和 Gemini 3 Pro 等模型构建的文本智能体和代码智能体。在“猫与狗”游戏中，采取稳妥行动可确保获得 5 分，而按某种信号约定行动只有在两个智能体选中同一种约定时才能获得 10 分。
与其他模型配对时，文本智能体的得分远低于稳妥行动所能获得的 5 分，每个智能体都认为自己的任意约定很自然。一段简短、共享的协调原则声明，被称为“有利于协调的定义”，使智能体能够可靠地选择稳妥行动，获得 5 分。
如果你的智能体必须与并非由你构建的智能体合作，就告诉它避免依赖未明说的约定，优先选择任何伙伴都能正确理解的行动。
原文
New Oxford paper finds that frontier LLM agents fail to coordinate with unfamiliar partners because each one treats its own made-up rules (arbitrary convention) as the obvious choice. Agents from different companies increasingly must work together with no agreed rules. Past studies used shared scaffolding or complex games that hide why coordination fails. They tested text and code agents built on models like GPT-5.2 and Gemini 3 Pro in 9 small cooperative games. In Cat & Dog, a safe move guarantees 5 points, while a signal convention pays 10 only if both agents pick the same one. Paired with other models, text agents scored well below that safe 5, each calling its own arbitrary convention natural. A short shared statement of coordination principles, called a Coordination-friendly definition, led agents to reliably take the safe 5. If your agent must work with agents you did not build, tell it to avoid unstated conventions and prefer actions any partner reads correctly.