arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2404.04285cs.CLcs.AI
收藏

MIMIR:面向领域专业知识个性化智能体调优的精简平台

MIMIR: A Streamlined Platform for Personalized Agent Tuning in Domain Expertise

Georgia Institute of Technology(佐治亚理工学院);Yale University(耶鲁大学);ETH Zurich(苏黎世联邦理工学院);MIT(麻省理工学院)

Chunyuan Deng, Xiangru Tang, Yilun Zhao, Hanming Wang, Haoran Wang, Wangchunshu Zhou, Arman Cohan, Mark Gerstein

AI总结:

针对开源LLM缺乏智能体调优数据的问题,本文提出Mimir端到端平台,整合私有与合规公开数据,支持个性化智能体调优及通用指令数据生成和一键微调。

摘要:

近来,大语言模型(LLM)已发展为交互式智能体,擅长在各类任务中进行规划、工具使用和任务执行。然而,若不进行专门的智能体调优,LLaMA等开源模型目前难以匹敌GPT-4的效率,尤其是在用于微调的智能体调优数据集稀缺的情况下。为此,我们推出Mimir:一个精简平台,提供可定制流水线,使用户能够大规模利用私有知识以及公开、合法合规的数据集,开展个性化智能体调优。此外,Mimir支持基于同一输入生成通用指令调优数据集。这种双重能力确保通过该平台开发的语言智能体同时具备特定智能体能力和通用能力。Mimir将这些功能整合为一个连贯的端到端平台,支持从上传个性化文件到一键智能体微调的全部流程。

英文摘要:

Recently, large language models (LLMs) have evolved into interactive agents, proficient in planning, tool use, and task execution across a wide variety of tasks. However, without specific agent tuning, open-source models like LLaMA currently struggle to match the efficiency of GPT- 4, particularly given the scarcity of agent-tuning datasets for fine-tuning. In response, we introduce \textsc{Mimir}: a streamlined platform offering a customizable pipeline that enables users to leverage both private knowledge and publicly available, legally compliant datasets at scale for \textbf{personalized agent tuning}. Additionally, \textsc{Mimir} supports the generation of general instruction-tuning datasets from the same input. This dual capability ensures that language agents developed through the platform possess both specific agent abilities and general competencies. \textsc{Mimir} integrates these features into a cohesive end-to-end platform, facilitating everything from the uploading of personalized files to one-click agent fine-tuning.

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。

↑