AgentArk:将多智能体智能蒸馏到单个LLM智能体中
AgentArk: Distilling Multi-Agent Intelligence into a Single LLM Agent
AI总结:
提出AgentArk框架,通过三种分层蒸馏策略将多智能体系统的交互动态蒸馏到单个模型权重中,使单个智能体具备多智能体的推理和自校正能力,同时保持计算效率。
摘要:
虽然大型语言模型(LLM)多智能体系统通过迭代辩论实现了卓越的推理性能,但实际部署受到高计算成本和错误传播的限制。本文提出AgentArk,一种新颖的框架,将多智能体动态蒸馏到单个模型的权重中,有效地将显式的测试时交互转化为隐式的模型能力。这使得单个智能体在保持计算效率的同时具备多智能体系统的智能。具体来说,我们研究了跨多种模型、任务、规模和场景的三种分层蒸馏策略:推理增强微调;基于轨迹的增强;以及过程感知蒸馏。通过将计算负担从推理转移到训练,蒸馏后的模型在保持单个智能体效率的同时,展现出多个智能体的强推理和自校正性能。它们还在各种推理任务中表现出增强的鲁棒性和泛化能力。我们希望这项工作能为未来高效且鲁棒的多智能体开发研究提供启示。我们的代码位于https://github.com/AIFrontierLab/AgentArk。
英文摘要:
While large language model (LLM) multi-agent systems achieve superior reasoning performance through iterative debate, practical deployment is limited by their high computational cost and error propagation. This paper proposes AgentArk, a novel framework to distill multi-agent dynamics into the weights of a single model, effectively transforming explicit test-time interactions into implicit model capabilities. This equips a single agent with the intelligence of multi-agent systems while remaining computationally efficient. Specifically, we investigate three hierarchical distillation strategies across various models, tasks, scaling, and scenarios: reasoning-enhanced fine-tuning; trajectory-based augmentation; and process-aware distillation. By shifting the burden of computation from inference to training, the distilled models preserve the efficiency of one agent while exhibiting strong reasoning and self-correction performance of multiple agents. They further demonstrate enhanced robustness and generalization across diverse reasoning tasks. We hope this work can shed light on future research on efficient and robust multi-agent development. Our code is at https://github.com/AIFrontierLab/AgentArk.
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。