Elon Musk 点赞 Kimi 技术报告,联合一作陈广宇 17 岁在读惠州贝赛思

Gate News 消息,3 月 17 日,月之暗面发布 Attention Residuals 技术报告,提出用注意力机制替代 Transformer 中固定的残差连接,在 Kimi Linear 48B 模型上可等效多用 25% 算力、推理延迟增加不到 2%。Elon Musk 昨晚在 X 发文「Impressive work from Kimi」,月之暗面官方今日在微博上回应「你的火箭造得也不错!」。

这条推文也将讨论引向论文的联合一作之一:陈广宇(英文名 Nathan),今年 17 岁,目前仍在读高中。论文另两名联合一作为 RoPE(旋转位置编码)提出者苏剑林,以及 Kimi Linear 第一作者张宇。陈广宇于 2025 年 11 月加入月之暗面,GitHub 上的 Flash Linear Attention 开源项目是他入门机器学习的起点。

陈广宇本人也在 X 上回应外界讨论,称这样一篇「算法和 infra codesign,同时实验和理论都有补充的 paper 是不太可能一个人写出来的」,Kimi 团队大家都有投入,Yu Zhang 与苏剑林也都是 equal contributor,提醒大家「不要相信谣言」。

陈广宇本人领英主页显示,其就读学校为惠州贝赛思(Basis International Park Lane Harbour)。Moonshot Academy 是 2025 年 3 月举办「Moonshot 48」高中生黑客松的主办方,陈广宇在该活动中获得冠军。

Disclaimer: The information on this page may come from third parties and does not represent the views or opinions of Gate. The content displayed on this page is for reference only and does not constitute any financial, investment, or legal advice. Gate does not guarantee the accuracy or completeness of the information and shall not be liable for any losses arising from the use of this information. Virtual asset investments carry high risks and are subject to significant price volatility. You may lose all of your invested principal. Please fully understand the relevant risks and make prudent decisions based on your own financial situation and risk tolerance. For details, please refer to Disclaimer.
Commento
0/400
Nessun commento