Résultats
100 articlesTrier
38:14EN
No Priors: AI, Machine Learning, Tech, & Startups· 18 sept. 2026Podcast · 2 voixDiffusion Will Win Inference: Stefano Ermon Bets Parallel Tokens Beat Custom SiliconAnime : Sarah Guo, Elad GilThe Stanford diffusion pioneer tells No Priors why a 50-person lab is training language models that generate many tokens at once — and why that maps to GPUs the way transformers once mapped to training.
36:19ENNo Priors: AI, Machine Learning, Tech, & Startups· 26 juin 2026Podcast · 2 voixThe Benchmark Grid Is Broken: Why Test-Time Compute Has Outrun How We Score ModelsAnime : Sarah Guo, Elad GilOpenAI researcher Noam Brown argues that static leaderboards hide how long a model is allowed to think — and that safety evals, research taste, and the release cycle have not caught up.
13:30EN
3Blue1Brown· 17 août 2025How AlphaGeometry Combines Logical Proof with Creative ConstructionAleph0Aleph0 explains how a symbolic geometry solver gets surprisingly far without AI, and how a language model supplies the auxiliary constructions that take it to Olympiad level.
42:09ENNo Priors: AI, Machine Learning, Tech, & Startups· 26 juin 2025Podcast · 2 voixAlphaEvolve found algorithms humans missed for 50 years — then sped up Gemini 23%Anime : Sarah Guo, Elad GilGoogle DeepMind’s Pushmeet Kohli and Matej Balog tell No Priors how an evolutionary coding agent searches algorithm space the way AlphaGo searched Go.
3:14:34EN
Lex Fridman· 14 juin 2025Podcast · 2 voixIf ten things make a problem hard, Terence Tao turns nine of them offInvité : Terence Tao · Anime : Lex FridmanThe Fields Medalist walks Lex Fridman from the Kakeya needle and Navier-Stokes through Lean, Perelman, and why today's AI still cannot smell a dead-end proof.
38:10ENNo Priors: AI, Machine Learning, Tech, & Startups· 1 mai 2025Podcast · 2 voixo3 should wait exactly as long as the answer takesAnime : 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.
2:49:35ZH
Zhang Xiaojun Podcast· 4 févr. 2025Podcast · 2 voix潘嘉怡解读 R1、o1 与 Kimi K1.5：强化学习如何教模型学会推理Invité : 潘嘉怡 · Anime : 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。
1:33:07ZHZhang Xiaojun Podcast· 17 août 2024Podcast · 2 voix广密复盘全球大模型这一年：Scaling Law、AGI 豪赌与中美格局Invité : 李广密 · Anime : 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。
34:09ENNo Priors: AI, Machine Learning, Tech, & Startups· 27 juin 2024Podcast · 2 voixFive Independent Teams Found the Same 10-to-1 Mix That Beats TransformersAnime : 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.
26:00ENNo Priors: AI, Machine Learning, Tech, & Startups· 11 juin 2024Podcast · 2 voixWhy a Zapier founder thinks AGI research has been stalled for four yearsAnime : Sarah Guo, Elad GilMike Knoop and François Chollet put more than a million dollars on beating ARC-AGI, the eval they say actually measures general intelligence.
2:15:39ENLex Fridman· 2 juin 2024Podcast · 2 voixRoman Yampolskiy Puts Superintelligent Ruin at 99.99 PercentInvité : Roman Yampolskiy · Anime : Lex FridmanThe AI safety researcher tells Lex Fridman that controlling AGI is as impossible as a perpetual motion machine, and that tests, open source, and regulation cannot close the gap.
2:50:44ENLex Fridman· 17 avr. 2024Podcast · 2 voixThe Pirahã Have No Word for OneInvité : Edward Gibson · Anime : Lex FridmanMIT psycholinguist Ted Gibson tells Lex Fridman why related words sit together, why language is not thought, and why large language models ace form while missing meaning.
2:47:17ENLex Fridman· 7 mars 2024Podcast · 2 voixIf You Want Human-Level AI, Abandon Generative ModelsInvité : Yann Lecun · Anime : Lex FridmanYann LeCun tells Lex Fridman why autoregressive LLMs cannot plan, remember, or understand the physical world — and why open source is the only democratic answer.
41:58ENNo Priors: AI, Machine Learning, Tech, & Startups· 2 nov. 2023Podcast · 2 voixReliability is the bottleneck. Autonomy is what would make AI feel alive.Anime : 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.
39:50ENNo Priors: AI, Machine Learning, Tech, & Startups· 20 juil. 2023Podcast · 2 voixWhy AI Video Is Still a Four-Second CorgiAnime : Sarah Guo, Elad GilMeta research director Devi Parikh explains why text-to-video has stalled on short animated images, and what it will take to get past that plateau.
3:11:35ENLex Fridman· 21 juin 2023Podcast · 2 voixSearch was always a hack: Marc Andreessen on AI, truth, and the next internetInvité : Marc Andreessen · Anime : Lex FridmanLex Fridman sits down with the Mosaic co-creator and Andreessen Horowitz founder to ask whether Google Search, the browser, and open-source AI survive the decade.
2:07:04ENLex Fridman· 28 mai 2023Podcast · 2 voixThe Ribosome Already Solved Digital ManufacturingInvité : Neil Gershenfeld · Anime : Lex FridmanNeil Gershenfeld tells Lex Fridman why Turing separated the head from the tape, how fab labs are the personal-computer moment for making things, and what it would take to grow a civilization from about twenty parts.
4:14:33ENLex Fridman· 9 mai 2023Podcast · 2 voixThe Next Token Is Not a Proof: Stephen Wolfram on ChatGPT and Computational TruthInvité : Stephen Wolfram · Anime : Lex FridmanIn his fourth conversation with Lex Fridman, Stephen Wolfram contrasts large language models with Wolfram|Alpha, then follows computational irreducibility from ChatGPT through observers, thermodynamics, and the three great theories of twentieth-century physics.
37:18ENNo Priors: AI, Machine Learning, Tech, & Startups· 4 mai 2023Podcast · 2 voixUse Retrieval for a Wi-Fi Password. Use Experts for a Company's Private Language.Anime : Sarah Guo, Elad GilGoogle Brain's Kelvin Guu on REALM, FLAN, model surgery, and why personalizing large models is still an unsolved interface problem.
2:27:58ENLex Fridman· 28 avr. 2023Podcast · 2 voixIt took Atlas fourteen years to walk the way Playter wantedInvité : Robert Playter · Anime : Lex FridmanBoston Dynamics CEO Robert Playter traces thirty years from a MIT basement pogo stick to Atlas, Spot, and Stretch — and why the company still designs machines that are allowed to fall.
2:30:28ENLex Fridman· 21 avr. 2023Podcast · 2 voixAI Only Has a Neocortex: Manolis Kellis on Evolutionary Baggage and Superintelligent PartnersInvité : Manolis Kellis · Anime : Lex FridmanOn his fifth Lex Fridman podcast, the MIT computational biologist argues that layered evolutionary history is what makes humans irreplaceable — and that alignment with AI has to run both ways.
53:37ENNo Priors: AI, Machine Learning, Tech, & Startups· 9 mars 2023Podcast · 2 voix30 models, 42 scenarios, 7 metrics: Percy Liang's public checkup for language modelsAnime : Sarah Guo, Elad GilA PubMed-only model once led the U.S. medical licensing exam — until a system roughly 200 times larger overtook it.
1:00:36ENNo Priors: AI, Machine Learning, Tech, & Startups· 2 févr. 2023Podcast · 2 voixForty games of Diplomacy, and nobody spotted the bot: Noam Brown’s 100,000x lessonAnime : Sarah Guo, Elad GilMeta’s Noam Brown tells Sarah Guo and Elad Gil how Cicero negotiated in English for 40 games on a tiny web community — and why thinking at decision time beat three years of scaling.
2:29:21ENLex Fridman· 6 déc. 2022Podcast · 2 voixThe Poker Bot That Never Tried to Read a SoulInvité : Noam Brown · Anime : Lex FridmanNoam Brown tells Lex Fridman how a Nash-equilibrium bot crushed heads-up no-limit Hold'em, how Pluribus did it six-handed, and why Diplomacy required an AI that could earn human trust.Charger plus24 sur 100 affichés