无需训练的智能调度器，让多个AI协作更高效可靠。
Training-Free Agentic AI: Probabilistic Control and Coordination in Multi-Agent LLM Systems
- 用概率判断选择最可能成功的AI助手，避免盲目尝试。
- 相比随机调度，减少28%的计算量、17%的调用次数和19%的耗时。
- 适合需要多AI协作但不想调参或训练的开发者使用。
多智能体大语言模型系统通过组合专用智能体实现复杂、长周期推理，但实际部署受限于低效的任务分配、噪声反馈和高交互成本。本文提出REDEREF，一种轻量级且无需训练的多智能体协作控制器，可提升递归委托中的任务路由效率。REDEREF集成四类机制：(i) 基于信念的委托，利用汤普森采样优先选择历史表现优异的智能体；(ii) 反思驱动的重路由，由校准后的LLM或程序化裁判触发；(iii) 依据证据选择而非输出平均；(iv) 带记忆感知先验，缓解冷启动问题。在多智能体分知识任务中，仅使用递归重试即达任务成功率上限，而基于信念的路由相较随机递归委托，降低28%的令牌消耗、17%的智能体调用次数、19%的完成时间，并在智能体或裁判退化时仍保持良好适应性。结果表明，简单可解释的贝叶斯控制策略可在不训练的前提下显著提升多智能体系统的效率与鲁棒性。
原文摘要 · Abstract (English)
Multi-agent large language model (LLM) systems enable complex, long-horizon reasoning by composing specialized agents, but practical deployment remains hindered by inefficient routing, noisy feedback, and high interaction cost. We introduce REDEREF, a lightweight and training-free controller for multi-agent LLM collaboration that improves routing efficiency during recursive delegation. REDEREF integrates (i) belief-guided delegation via Thompson sampling to prioritize agents with historically positive marginal contributions, (ii) reflection-driven re-routing using a calibrated LLM or programmatic judge, (iii) evidence-based selection rather than output averaging, and (iv) memory-aware priors to reduce cold-start inefficiency. Across multi-agent split-knowledge tasks, we show that while recursive retry alone saturates task success, belief-guided routing reduces token usage by 28%, agent calls by 17%, and time-to-success by 19% compared to random recursive delegation, and adapts gracefully under agent or judge degradation. These results demonstrate that simple, interpretable probabilistic control can meaningfully improve the efficiency and robustness of multi-agent LLM systems without training or fine-tuning.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。