DPBench: 多智能体LLM在同时资源竞争下的协调结构决定因素
DPBench: Structural Determinants of Multi-Agent LLM Coordination Under Simultaneous Resource Contention
AI总结：
本文提出DPBench，用于评估多智能体系统中协调性能的基准测试，通过分析不同协议、通信结构和群体规模对协调成功或失败的影响，揭示了多智能体LLM在资源竞争中的协调机制。
摘要：
我们提出了DPBench，一个用于评估多智能体系统中协调性能的基准测试，该测试基于大型语言模型构建。现有基准测试在固定协议下衡量任务级的成功率；然而，协调成功或失败的结构条件尚未被明确刻画。DPBench将哲学家就餐问题改编为受控测试平台，其中动作协议、通信结构和群体规模可独立变化。我们评估了六个智能体：GPT-5.2、Claude Opus 4.5、Grok 4.1、Gemini 2.5 Flash、Llama 4 Maverick以及一个均匀随机基线。在N=5的同时动作下，默认提示中，GPT-5.2的死锁率为25.0%（95% Wilson置信区间[11.2, 46.9]），而Gemini 2.5 Flash的死锁率为90.0%（[74.4, 96.5]）；顺序动作被六个智能体中的四个解决。在固定模型为Gemini 2.5 Flash的情况下，三个协议变量将死锁率从90%降低到置信区间接近零：三次预承诺通信（0.0% vs. 单次通信86.7%）、提示中包含经典并发原语（资源排序和对称打破的0.0% vs. 最小提示的100%）或将群体从N=5扩大到N=10（90.0%到10.0%）。单次通信和过去时间步的记忆在我们运行的样本量下不会改变死锁率。是否同一个模型协调或死锁由协议决定，而不是模型的能力。
英文摘要：
We present DPBench, a benchmark for evaluating coordination in multi-agent systems built from large language models. Existing benchmarks measure task-level success under a fixed protocol; the structural conditions under which coordination succeeds or fails at all have not been characterised. DPBench adapts the Dining Philosophers problem into a controlled testbed where the action protocol, the communication structure, and the group size each vary independently. We evaluate six agents: GPT-5.2, Claude Opus 4.5, Grok 4.1, Gemini 2.5 Flash, Llama 4 Maverick, and a uniform-random baseline. Under simultaneous action at N=5 with the default prompt, deadlock ranges from 25.0% (95% Wilson CI [11.2, 46.9]) for GPT-5.2 to 90.0% [74.4, 96.5] for Gemini 2.5 Flash; sequential action is solved by four of the six. Holding the model fixed at Gemini 2.5 Flash, three protocol variables drive deadlock from 90% to within CI of zero: three rounds of pre-commitment communication (0.0% vs. single-round 86.7%), a prompt encoding a classical concurrency primitive (0.0% for resource-ordering and symmetry-breaking, against 100% for the minimal prompt), or doubling the group from N=5 to N=10 (90.0% to 10.0%). Single-round messaging and memory of past timesteps do not change the rate at the sample size we ran. Whether the same model coordinates or deadlocks is determined by the protocol, not by the model's capability.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
