筛选结果
66 篇排序
1:30:54ENHumanoids Outran Usain Bolt. DeepMind Still Scores Robotics as GPT-2主持 Nathan LabenzGoogle DeepMind's Gemini Robotics lead on why sprinting is easy, frying an egg is hard, and one brain should drive any body.
1:01:14ENThe Model Eats the Scaffolding: How Google Plans to Ship Agents Everywhere主持 Nathan LabenzThree years after the leaked “no moat” memo, DeepMind’s Gemini leads walk through 3.5 Flash, Omni, and the shared agent harness now running through Google’s products.
1:57:29ENFrontier vision models score 12.5%: Joseph Nelson on the real waterline of computer vision主持 Nathan LabenzThe Roboflow CEO tells Nathan Labenz why vision still lags language, how to distill SAM 3 into an edge model for Wimbledon, and why the United States has almost never led open-source vision.
1:01:38EN13-Hour Coffee and 2x Throughput: Physical Intelligence Crosses the Deployment LineKarol Hausman and Tobi Springenberg tell Sequoia’s Training Data how a general robot brain, plus real-world RL, turned coffee, laundry, and chocolate boxes into jobs that actually run.
39:16EN500 Gloves, 10 Million Trajectories, and a $10,000 Home Robot主持 Sarah Guo, Elad GilSunday Robotics co-founders Tony Zhao and Cheng Chi map the gap between a scalable imitation-learning recipe and a machine that actually does the dishes.
43:38ENA 2 A.M. Codename and a Face That Finally Looked Like YouGoogle's product and model leads walk Sequoia through how Nano Banana made single-photo character consistency feel real — and why fun, craft, and human evals did as much work as scale.
1:00:26ENSeven Million Videos a Day, and Most Return Visits Are Spent CreatingOpenAI's Sora 2 leads walk through diffusion transformers, physics that rebounds off the backboard, and a feed designed against mindless scrolling.
42:20ENA Name Invented at 2:30 a.m. Put Gemini on Top of the App Store主持 Jacob EffronGoogle's Nicole Brichtova and Oliver Wang on Nano Banana, why image models got smart when language models did, and why a prompt still is not a finished product.
1:55:02ENA hundred experiments a day, and the valley of death in materials主持 Nathan LabenzRadical AI’s co-founders tell Nathan Labenz how a closed loop of frontier models and a self-driving lab is meant to collapse a forty-year materials cycle into less than a decade.
2:28:14EN → 简体中文自然有结构:Demis Hassabis 谈可学习的现实与 AGI嘉宾 Demis Hassabis · 主持 Lex Fridman谷歌 DeepMind 的 CEO 认为,经过进化塑造的系统可以用经典学习方法来建模,而可交互的世界模型正处在通往 AGI 的路上。
35:47ENFei-Fei Li’s new bet: AI is incomplete without spatial intelligence主持 Sarah Guo, Elad GilThe ImageNet pioneer tells Sarah Guo and Elad Gil why World Labs is building 3D generative foundation models, and why language was only one piece of the puzzle.
53:51ENVeo 3 Took First Place by Beating Veo 2. Google Labs Says Chapter Two Just StartedThree Google Labs leads walk through generative cameras, background computer-use agents, and notebooks that adapt knowledge for an audience of one.
1:19:10ENA Robot-Arm Shot That Took Months Now Lives in a Prompt主持 Nathan LabenzLuma Labs’ Amit Jain and Jiaming Song on Ray 2, few-shot camera concepts, and why they think video models sit on the path to multimodal AGI.
38:10ENo3 should wait exactly as long as the answer takes主持 Sarah Guo, Elad GilOpenAI researchers Eric Mitchell and Brandon McKinzie tell No Priors why o3's leap is a chain of tools during reasoning — and why you should send the same prompt a hundred times.
40:54ENDiscovery rates collapsed tenfold. KoBold still found a 5% copper deposit主持 Sarah Guo, Elad GilJosh Goldman explains why KoBold Metals treats mineral exploration as an information problem, pairing AI with century-old maps to hunt lithium and copper.
1:43:54ENA Year to Grow a Brain Organoid, an Hour to Simulate It主持 Nathan LabenzStanford postdoc Siyu He tells Nathan Labenz how Squidiff and CORAL move single-cell and spatial experiments into the computer — and where the models still cannot go.
35:14ENRobots Have No Wikipedia. Chelsea Finn Is Starting With Three Buildings.主持 Sarah Guo, Elad GilPhysical Intelligence’s cofounder tells No Priors why a generalist robot model has to be trained across embodiments, why open-sourcing hardware is a bet against the problem going unsolved, and why humanoids are a little overrated.
3:53:49ENWith One-Twelfth the Data, an Open-Source Vision Model Beat GPT-4o主持 Nathan LabenzWill Hardman of Veratai walks Nathan Labenz through two years of vision-language models, from ViT and CLIP to InternVL, BLINK, and Transfusion.
1:07:32ENo1 Is a New Generation — and the Day AGI Arrives Will Feel Boring主持 Jacob EffronOpenAI’s former chief research officer Bob McGrew on why o1 is a 100x leap, why agents stall on reliability, and why AGI may land as office tedium rather than a single dramatic moment.
55:16ENGemini’s API Grew 14× in Six Months. Google’s PMs Explain the Catch-Up.主持 Nathan LabenzLogan Kilpatrick and Shrestha Basu Mallick walk through search grounding, Gemini Flash’s price-latency quadrant, and why a near-tied leaderboard still leaves Google third in most developers’ queues.
54:45ENA Few Megabytes of Doom and a Few Gigabytes of Model: Why Runway Bets on Generalization主持 Nathan LabenzCTO Anastasis Germanidis tells The Cognitive Revolution why video is a fuller record of the world than text, why physics shows up without being programmed in, and how a small lab competes with unlimited compute.
2:04:15ENA parrot cannot paraphrase — and Claude cannot be talked into harm主持 Nathan LabenzRobert Wright of the Nonzero podcast interviews Nathan Labenz on whether large language models understand anything, what Anthropic’s 10 million features actually show, and why open weights collide with U.S.–China policy.
1:16:22ENHe Spent 20 Hours Crafting a Prompt. DSPy Beat Him Anyway主持 Nathan LabenzSander Schulhoff walks through The Prompt Report — a 78-page survey of what actually works when you prompt large language models.
1:37:07ENA Text-Only AI Outdiagnosed GPs. Google Still Won't Ship It.主持 Nathan LabenzGoogle researchers Vivek Natarajan and Khaled Saab walk Nathan Labenz through five medical-AI papers, from radiology reports to an AI doctor that still has to earn trust.
加载更多已显示 24 篇,共 66 篇









