Ergebnisse
7 ArtikelSortieren
2:36:13ZH松琳解读 DeepSeek、Kimi 与 MiniMax 的注意力新论文Gast: Songlin Yang(松琳) · Moderation: 张小珺从动态稀疏注意力到线性注意力混合架构:三条长上下文路线如何兼顾训练、推理与硬件效率
44:16ENA perfect demo is ten years from a product you can pay forModeration: Sarah Guo, Elad GilAndrej Karpathy uses Waymo's decade-long gap, Tesla's software bet, and a new AI course to argue that autonomous intelligence will stall at product and globalization, not at the first impressive clip.
34:09ENFive Independent Teams Found the Same 10-to-1 Mix That Beats TransformersModeration: Sarah Guo, Elad GilThe authors of S4 and Mamba argue that mostly state-space layers plus a dash of attention is the architecture that actually works—and they are shipping it first as real-time speech.
41:58ENReliability is the bottleneck. Autonomy is what would make AI feel alive.Moderation: Sarah Guo, Elad GilIlya Sutskever talks with Sarah Guo and Elad Gil about AlexNet, OpenAI’s capped-profit bet, why smaller models mainly lose reliability, and the line he draws for digital life.
35:24ENThe Author of Attention Is All You Need Now Compiles mRNA Like BytecodeModeration: Sarah Guo, Elad GilJakob Uszkoreit on why Transformers won, why models spend compute on the wrong things, and why drug discovery has to become generation rather than screening.
35:36ENAGI as a side effect: Noam Shazeer on Transformers, Character.AI, and two-hour chatsModeration: Sarah Guo, Elad GilIn April 2023, the Transformer co-author told No Priors how a 22-person startup turned open-ended chat into a consumer product — and treated AGI as a byproduct of making the model smarter.
1:46:01ENThe Night AlphaStar Beat TLO, the Team Went to a BarGast: Oriol Vinyals · Moderation: Lex FridmanOriol Vinyals on how DeepMind turned StarCraft II into a sequence problem, built a league of agents, and still refuses to call the game solved.








