有人说前面提到月之暗面的研究结果早有人发过类似论文,贡献评估还是得过会议双盲审才可靠;不管怎样,即使是重新发现也引人注目

本帖于 2026-03-16 21:42:30 时间, 由普通用户 米汤 编辑

https://x.com/behrouz_ali/status/2033581834953453853

https://bbs.wenxuecity.com/znjy/7800831.html

also:

Zhilin Yang (founder Moonshot) is speaking at GTC on scaling Kimi 2.5.

Time: Tuesday, March 17, 11:00–11:40 a.m. PDT (GTC 2026 in San Jose).

He is one of the coauthors of the original T5 paper, “Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer”, cmu PhD '19

https://x.com/bookwormengr/status/2033527837333086559

https://kimiyoung.github.io/

请您先登陆,再发跟帖!