arXiv cs.AI· Yuqi Jia, Yunjie Tian, Jiayu Zheng, Yijun Liang, Jingchen Sun, Mengyu Zhao, Di Fu·· 1 天前AI 评分58
Axion 1.0 发布：简化训练流程的开源智能体基础模型
Axion 1.0: Open Agent Foundation Models, Simplified
AI 导读
Axion 1.0 是一个完全开源的智能体基础模型，训练只经过两个轻量监督微调阶段和强化学习，不含专门的中期训练、DPO 或 on-policy 蒸馏。第一阶段通过模型感知的数据构建强化自包含推理，第二阶段用可扩展流水线构建基于真实数据源和确定性计算的可自动验证工具任务，强化学习再提升长程推理与工具交互能力。
正文 · 原文
Abstract:We introduce Axion 1.0, a fully open agent foundation model that combines strong and broad agentic capabilities with a remarkably simple training recipe. Axion 1.0 is trained in only three stages: two lightweight supervised fine-tuning stages followed by reinforcement learning, without dedicated mid-training, direct preference optimization (DPO), or on-policy distillation (OPD). The first SFT stage develops strong self-contained reasoning through model-aware data construction, where we synthesize challenging problems and adversarially strengthen them based on the base model's own successful solution strategies. The second stage extends these capabilities to tool-dependent reasoning through a scalable pipeline for constructing diverse and automatically verifiable tasks grounded in real-world data sources and deterministic computation. Finally, reinforcement learning further strengthens long-horizon reasoning and interaction with external tools and environments. Despite this compact training recipe, Axion 1.0 achieves strong performance across a broad suite of agentic benchmarks and is competitive with leading commercial models on several of them. Equally importantly, Axion 1.0 is open beyond model weights: we release the complete recipe, including data construction, generation, filtering, verification, prompts, and training code, enabling the entire development process to be reproduced and extended. Axion 1.0 demonstrates that strong agent foundation models can be both open and simple, offering a practical and reproducible path toward advancing open agentic intelligence.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.03233 [cs.AI] |
| (or arXiv:2611.03233v3 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2611.03233 arXiv-issued DOI via DataCite |
Submission history
From: Yuqi Jia [view email]
[v1]
Mon, 2 Nov 2026 14:48:57 UTC (268 KB)
[v2]
Tue, 3 Nov 2026 10:54:02 UTC (276 KB)
[v3]
Wed, 11 Nov 2026 15:24:40 UTC (276 KB)
来源：arXiv cs.AI · arxiv.org