Transistor (https://transistor.fm) https://feeds.transistor.fm/jiggly-jar Testing © 2026 Shaoqing Tan 4a592eba-7734-51e9-862e-8ceb3afc55d5 yes en Thu, 14 May 2026 15:45:46 -0700 Thu, 14 May 2026 15:46:13 -0700 https://img.transistorcdn.com/OwmSD54nWl4YFcmlfz-RAFxALTAKgjo-Wx9-nHd6Vk8/rs:fill:0:0:1/w:1400/h:1400/q:60/mb:500000/aHR0cHM6Ly9pbWct/dXBsb2FkLXByb2R1/Y3Rpb24udHJhbnNp/c3Rvci5mbS8xMGQ3/ZjVhZjJjNTFjMTll/YmM4YTZjMDAxYjRk/Y2YxZi5qcGVn.jpg episodic Shaoqing Tan Testing Testing. Shaoqing Tan No No AnyFlow: Any-Step Video Diffusion with Flow Map Distillation full 547a85a1-e6e6-4c41-a816-e9ec3170729e https://share.transistor.fm/s/ba279204 Thu, 14 May 2026 15:45:46 -0700 Shaoqing Tan Shaoqing Tan 784 Introduces AnyFlow, the first any-step video diffusion framework using flow maps that allows a single model to adapt to arbitrary inference steps for high-quality video generation, enabling efficient predictive video modeling for world models. Introduces AnyFlow, the first any-step video diffusion framework using flow maps that allows a single model to adapt to arbitrary inference steps for high-quality video generation, enabling efficient predictive video modeling for world models. No Robotics: Endgame – NVIDIA's Roadmap to Physical AGI full 333cb8a4-bf35-49fc-b720-1694c4b7a955 https://share.transistor.fm/s/8012d82d Thu, 14 May 2026 15:41:21 -0700 Shaoqing Tan Shaoqing Tan 1885 A comprehensive talk outlining NVIDIA's roadmap to Physical AGI, proposing video world models as a second pretraining stage, introducing World Action Models (WAM) for joint video-action prediction, and describing systems like EgoScale and DreamDojo for robotics scaling. A comprehensive talk outlining NVIDIA's roadmap to Physical AGI, proposing video world models as a second pretraining stage, introducing World Action Models (WAM) for joint video-action prediction, and describing systems like EgoScale and DreamDojo for ro No DreamZero: World Action Models as Zero-Shot Robotic Policies full 64ad86ee-dfc3-4c28-90aa-b54fe5855941 https://share.transistor.fm/s/0c7d8cf4 Sun, 22 Mar 2026 15:26:57 -0700 Shaoqing Tan Shaoqing Tan 1636 Introduces World Action Models (WAMs) that jointly predict video and actions via a 14B-parameter autoregressive diffusion model, enabling state-of-the-art zero-shot generalization on manipulation benchmarks like MolmoSpaces and RoboArena through emergent internal world modeling. Introduces World Action Models (WAMs) that jointly predict video and actions via a 14B-parameter autoregressive diffusion model, enabling state-of-the-art zero-shot generalization on manipulation benchmarks like MolmoSpaces and RoboArena through emergent No Test Episode 1 full 64402402-692a-406d-8a89-2ed5a880dff0 https://share.transistor.fm/s/e86d063a Sat, 21 Mar 2026 17:48:42 -0700 Shaoqing Tan Shaoqing Tan 1140 No