Axion 1.0开源智能体模型训练方案
AI 总结
Axion 1.0是一款开源智能体基础模型,训练分为三个阶段:两个轻量级监督微调阶段,随后进行强化学习,没有专门的中期训练、直接偏好优化或在线策略蒸馏。此次发布包括数据构造、生成、筛选、验证、提示词和训练代码。论文报告称,122B模型在限制网站访问的完整Humanity’s Last Exam上取得59.0%,在FrontierScience Research上取得66.7%。
深度见解
此次发布不仅提供模型权重,还公开了数据构造、验证和训练的完整方案。论文表示,这些材料使整个开发流程能够被复现和扩展,为开放智能体AI研究提供了一条实用且可复现的路径。
Computer Science > Artificial Intelligence
arXiv:2611.03233 (cs)
[Submitted on 2 Nov 2026 (v1), last revised 3 Nov 2026 (this version, v2)]
Title:Axion 1.0: Open Agent Foundation Models, Simplified
Authors:Yuqi Jia, Yunjie Tian, Jiayu Zheng, Yijun Liang, Jingchen Sun, Mengyu Zhao, Di Fu
View a PDF of the paper titled Axion 1.0: Open Agent Foundation Models, Simplified, by Yuqi Jia and 6 other authors
#Axion 1.0#ByteDance#Qwen3.6-35B-A3B#Qwen3.5-122B-A10B#Duke University#University of Maryland, College Park#University at Buffalo