PersonalAILab/AFM-CodeAgent-SFT-Dataset
收藏Hugging Face2025-08-20 更新2025-09-13 收录
下载链接：
https://hf-mirror.com/datasets/PersonalAILab/AFM-CodeAgent-SFT-Dataset
下载链接
链接失效反馈官方服务：
资源简介：
该数据集是针对Agent Foundation Models (AFMs)训练设计的核心训练数据，目的是激发大型语言模型中的端到端多智能体推理能力。数据集基于Chain-of-Agents (CoA)范式，包含从先进多智能体系统中提取的高质量轨迹的SFT数据，以及用于增强模型在CoA推理中的鲁棒性和性能的RL数据。
This dataset is designed as the core training data for Agent Foundation Models (AFMs) to elicit end-to-end multi-agent reasoning capabilities in large language models. Built on the Chain-of-Agents (CoA) paradigm, it includes SFT data extracted from advanced multi-agent systems and RL data to enhance the models robustness and performance in CoA reasoning.
提供机构：
PersonalAILab
