Ergebnisse
7 ArtikelSortieren
2:51:18EN
Zhang Xiaojun Podcast· 15. Okt. 2026Podcast · 2 StimmenLindon Gao on Building Dyna Robotics and Finding a Path Through the FogGast: Lindon Gao · Moderation: 张小珺Dyna Robotics CEO Lindon Gao explains why his company builds intelligence through real workflows, how hardware and research must evolve together, and how a founder can find clarity by letting go of the need to be right.
2:59:03ENZhang Xiaojun Podcast· 28. Juli 2026Podcast · 2 StimmenYou Kaichao on How vLLM Grew from Open-Source Project to AI Infrastructure CompanyGast: You Kaichao · Moderation: 张小珺Inferact co-founder You Kaichao traces vLLM’s path from Berkeley research to a company, then explains why model, inference, hardware and agent design increasingly have to move together.
3:46:12ENZhang Xiaojun Podcast· 16. Juli 2026Podcast · 2 StimmenKay Ke on Physical Intelligence, Robot Learning, and the Future of RoboticsGast: Kay Ke · Moderation: 张小珺Physical Intelligence researcher Kay Ke traces the field's research lineages and explains Pi's progression from π₀ to π*₀.₆. She also reflects on education, fiction, ambition, and what it means to build robots that work beyond a demo.
2:36:13ZHZhang Xiaojun Podcast· 24. Feb. 2025Podcast · 2 Stimmen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
2:49:35ZHZhang Xiaojun Podcast· 4. Feb. 2025Podcast · 2 Stimmen潘嘉怡解读 R1、o1 与 Kimi K1.5：强化学习如何教模型学会推理Gast: 潘嘉怡 · Moderation: 张小珺伯克利博士生潘嘉怡带读 DeepSeek、OpenAI 与 Kimi 的推理模型报告，拆解长思维链、GRPO、奖励设计与蒸馏。R1-Zero 的“顿悟时刻”展示了推理策略如何在强化学习中涌现。
1:27:13ZHZhang Xiaojun Podcast· 29. Dez. 2024Podcast · 2 Stimmen李想：汽车将从交通工具进化为空间机器人Gast: 李想 · Moderation: 张小珺理想汽车 CEO 李想谈公司为何从造车走向人工智能：基座模型是 AI 时代的操作系统，汽车是进入物理世界的重要载体。对他而言，技术路线最终要落实为更好的用户体验、能力与长期价值。
1:33:07ZHZhang Xiaojun Podcast· 17. Aug. 2024Podcast · 2 Stimmen广密复盘全球大模型这一年：Scaling Law、AGI 豪赌与中美格局Gast: 李广密 · Moderation: 张小珺拾象创始人李广密从硅谷一线观察出发，回顾 OpenAI、Anthropic、Google 与中国模型公司的竞争。他认为，模型能力与成本共同决定新一轮应用爆发，而巨额投入押注的是智能能否持续扩展以及 AI 对经济和产业的长期重构。