arXiv · cs.AI · Jingnan Zheng, Cheng Jin, Ming Zhang, Huan Sun, Zhongtian Ma, Kaisen Yang, Ziyuan Zhou, Tong Wu, Dongcheng Zhang, Yi Zhang, Xiao Wang, Han Qi, Shu Yu, Ermo Hua, Zongji Yu, Chaozhuo Li, Xiaoyu Wen, Zhekai Chen, Jiawei Zhou, Sirui Chen, Ning Ding, An Zhang, Tat-Seng Chua, Xiangnan He, Xia Hu, Bowen Zhou, Qiaosheng Zhang, Youbang Sun, Xiang Wang, Zhongyu Huang, Chaochao Lu · 10 小时前模型评分42
Safin-1.5：以安全原生架构与可验证推理实现递归安全提升
Safin-1.5: Safety-Native Architecture and Verifiable Reasoning for Recursive Safety Improvement
导读
Safin-1.5 提出安全原生架构与递归安全提升方法，通过“发现—评估—修复”循环持续修补风险，并只在安全提升且通用能力不降、过度拒答受限时接受更新。模型还配备推理时形式化推理框架，结合符号推理与验证生成可检查证明。实验显示其在安全与推理性能上超过前沿模型。
分类置信度
中 · 待人工复核
- 只有单一媒体来源，需要人工复核。
来源级别：媒体与个人
事件类型：研究发布
证据摘录
- “We introduce Safin-1.5, a model built on a safety architecture and trained through recursive safety improvement, with an inference-time harness that extends its capabilities to formal reasoning.”
来源：arXiv · cs.AI · arxiv.org