难度感知拓扑选择：多智能体代码生成的自适应协作
原标题：Learning How Much to Collaborate: Difficulty-Aware Topology Selection for Multi-Agent Code Generation
AI 摘要
该研究指出多智能体代码生成系统通常固定使用单一通信拓扑，但实验发现层级协作相对单智能体的优势随问题难度从2.4分（简单）升至21.1分（困难），而token成本始终约为10倍。作者提出Difficulty-Aware Topology Selector（Dats），用图网络预测各拓扑解题概率并结合成本选择最优拓扑，在预算匹配协议下以40%的层级成本达到77.7% pass@1，优于始终层级（73.6%）和最强学习基线（74.3%）。
正文节选
Learning How Much to Collaborate: Difficulty-Aware Topology Selection for Multi-Agent Code Generation Abstract Multi-agent systems for code generation are deployed with a single communication topology, chosen once for every problem. This is the wrong granularity. Evaluating five topologies on 614 problems from APPS, HumanEval+ and LiveCodeBench, we find that the advantage of hierarchical collaboration over a single agent grows from 2.4 points of pass@1 on the easiest third of problems to 21.1 po