arXiv cs.AI· Adrian Hayler, Shashank Reddy Chirra, Andrei Lupu, Johannes Forkel, Bidipta Sarkar, Siheng Feng, Jakob Foerster·· 3 小时前AI 评分50
研究：LLM 智能体间的零样本协调仍面临挑战
Zero-Shot Coordination among LLM Agents
AI 导读
一项 arXiv 研究考察了独立开发的 LLM 智能体在测试时进行零样本协调（ZSC）的能力。在使用简单通用脚手架和极简环境的受控实验中，前沿 LLM 智能体仍难以协调，主因是对协调问题理解有限、对伙伴信念的推理较弱，且未能有效利用语义信息。作者提出协调友好定义（CFDs）作为实现 LLM 智能体间稳健协调的原则性方法，并可自动发现、免去人工设计，代码已开源。
正文 · 原文
Abstract:We study zero-shot coordination (ZSC), where independently developed agents must coordinate at test time. While ZSC has been well studied in the RL literature, far less is known about the performance of LLM agents despite their increasing deployment in such settings. Existing work on LLM coordination often relies on specialised scaffolding that independently developed agents are unlikely to share in practice. Moreover, evaluations on complex environments (e.g., Hanabi) make it difficult to pinpoint the sources of coordination failure. In contrast, we focus on simple, general-purpose scaffolds in minimal environments designed to isolate specific coordination challenges. Our results show that even in these controlled settings, frontier LLM agents struggle to coordinate, largely due to limited understanding of the coordination problem and weak reasoning about their partner's beliefs. While LLMs introduce semantic information as an additional axis for coordination, they nonetheless fail to exploit this structure effectively. Towards this, we propose Coordination-friendly definitions (CFDs) as a principled approach for enabling robust coordination among LLM agents. Finally, we show that CFDs can be discovered automatically, removing the need for manual engineering. Code is available at this https URL.
| Comments: | NeurIPS 2026 Poster. ICLR 26 MALGAI Workshop Oral. The first two authors contributed equally |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.06409 [cs.LG] |
| (or arXiv:2611.06409v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2611.06409 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Shashank Reddy Chirra [view email]
[v1]
Wed, 4 Nov 2026 12:12:11 UTC (1,680 KB)
来源：arXiv cs.AI · arxiv.org