发表机构
Sber AI; Moscow Institute of Physics and Technology; National University of Science and Technology MISIS; Skolkovo Institute of Science and Technology; Artificial Intelligence Research Institute（Sber AI; 莫斯科物理技术学院; 莫斯科国立科学技术大学MISIS; 斯科尔科沃科学技术研究所; 人工智能研究所）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究提出LLM Agents Factory框架，通过检索2万余个预设智能体配置文件按需构建领域特定智能体，在多基准测试中实现高准确率与低推理成本，为工业应用提供动态智能体生成的替代方案。
AI 中文摘要
大语言模型（LLM）智能体通过将问题分解为角色专门化的行为来提升任务性能。然而，它们的实际部署常受限于为每个用户请求动态设计智能体带来的计算成本与不稳定性。为解决该问题，我们提出LLM Agents Factory，这是一种基于检索的框架，可利用超过20000个预设智能体配置文件的基础，按需构建领域特定且基于维基百科的智能体。我们的框架支持两种模式：（1）通过语义搜索进行智能体配置文件检索；（2）蒸馏为针对直接生成智能体而微调的紧凑模型。在单智能体场景下于MMLU、BIG-bench和BIG-bench Hard上开展的实验表明，我们基于检索的智能体构建方法在准确率上优于非智能体基线，且以120B主干模型实现的AutoGen生成质量相当，同时推理成本显著更低。本研究揭示，从结构化智能体仓库进行检索，为动态智能体生成提供了一种高性价比、准确且可控的替代方案，可满足工业应用的严格需求。我们在该https网址提供了实现代码与智能体库。
英文摘要
Large language model (LLM) agents improve task performance by decomposing problems into role-specialized behaviors. However, their practical deployment is often limited by the computational cost and instability associated with the on-the-fly agent design for each user request. To address this, we present LLM Agents Factory, a retrieval-based framework that constructs domain-specific and Wikipedia-grounded agents on demand using a base of over 20K predetermined agent profiles. Our framework supports two modes: (1) agent profile retrieval via semantic search and (2) distillation into a compact model fine-tuned for direct agent generation. Experiments on MMLU, BIG-bench, and BIG-bench Hard in a single-agent scenario demonstrate that our retrieval-based agent construction surpasses non-agent baselines in accuracy while matching AutoGen generation quality with a 120B backbone at a substantially lower inference cost. Our work reveals that retrieval from a structured agent repository provides a cost-efficient, accurate, and controllable alternative to dynamic agent generation, responding to the strict demands of industrial applications. We provide the implementation code and the agent base in https://huggingface.co/frontier-ai/llm-agent-factory.
Comments7 pages, 1 figure, SIGIR 2026
