Resultados
381 artículosOrdenar
46:09EN
No Priors: AI, Machine Learning, Tech, & Startups· 1 ago 2024Pódcast · 2 vocesOriol Vinyals: The Days of Burning 90% of Compute on Pretraining Are NumberedPresenta: Sarah Guo, Elad GilGoogle DeepMind’s VP of research on Gemini, long context, the coming shift from pretraining to RL, and why AGI may never have a finish line.
13:40EN
Matthew Berman· 29 jul 2024Testing Llama 3.1 8B: Benchmark Gains and Mixed Practical ResultsMatthew Berman · Forward FutureMatthew Berman tests coding, reasoning and other prompts on the small model and finds several failures despite stronger reported benchmarks.
10:02ENMatthew Berman· 24 jul 2024Testing Llama 3.1 405B on Coding, Arithmetic and ReasoningMatthew Berman · Forward FutureMatthew Berman tries Meta's 2024 open-weight model on an informal set of prompts and demonstrates a sponsored dataset workflow.
15:50ENMatthew Berman· 23 jul 2024Llama 3.1: Meta’s 405B Model and the Open-Weight EcosystemMatthew Berman · Forward FutureMatthew Berman reviews Meta’s July 2024 release, the smaller models, reported benchmarks, synthetic data and the proposed Llama Stack.
0:58ENMatthew Berman· 11 jul 2024NVIDIA NIM: Containerized AI Inference MicroservicesMatthew Berman · Forward FutureA sponsored overview of NVIDIA’s approach to packaging model inference for local and cloud applications.
18:57ENMatthew Berman· 9 jul 2024AI News in July 2024: Small Models, Voice Tools and OpenAI SecurityMatthew Berman, Andrej Karpathy · Forward FutureMatthew Berman reviews AI product announcements, model evaluation and security reporting, and shares his predictions about the future of computing.
1:42:16EN
Cognitive Revolution "How AI Changes Everything"· 4 jul 2024Pódcast267 Papers in Seven Months: Albert Gu Explains the State Space Model RevolutionPresenta: Nathan LabenzThe Mamba co-author walks through five years of recurrent-model work, why selective compression beats a growing KV cache, and what Mamba 2 trades away to train faster.
13:30ENMatthew Berman· 2 jul 2024AI News in July 2024: Smart Glasses, Robot Skin and Model RoutingMatthew Berman, Mark Zuckerberg · Forward FutureMatthew Berman surveys early plans for AI glasses, robotics demonstrations, generative video and RouteLLM's model-routing results.
9:06ENMatthew Berman· 28 jun 2024Testing Nemotron-4 340B: Synthetic Data and a Handful of Reasoning TasksMatthew Berman · Matthew BermanMatthew Berman reviews NVIDIA’s 2024 model family for synthetic data and tries a small set of coding, arithmetic and logic prompts.
34:09ENNo Priors: AI, Machine Learning, Tech, & Startups· 27 jun 2024Pódcast · 2 vocesFive Independent Teams Found the Same 10-to-1 Mix That Beats TransformersPresenta: 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.
1:54:01ENCognitive Revolution "How AI Changes Everything"· 26 jun 2024PódcastA Tiny Network Grew a Salt Crystal the Training Data Never ShowedPresenta: Nathan LabenzTim Duignan trained tens of thousands of parameters on salt water. The model invented a crystal — and a continuum solvent theory he had spent a PhD failing to write by hand.
58:18EN
20VC with Harry Stebbings· 24 jun 2024Pódcast · 2 vocesDavid Luan on Model Scaling, AI Agents and Competition Across the Chip StackInvitado: David Luan · Presenta: Harry StebbingsThe Adept co-founder explains his views on compute, vertical integration, enterprise agents and why human interaction should shape AI product design.
12:55ENMatthew Berman· 23 jun 2024Testing Mixture-of-Agents on Coding, Arithmetic and ReasoningMatthew Berman · Forward FutureMatthew Berman tries a multi-model setup on his informal LLM rubric and finds strong individual answers alongside coding and counting failures.
1:37:07ENCognitive Revolution "How AI Changes Everything"· 22 jun 2024PódcastA Text-Only AI Outdiagnosed GPs. Google Still Won't Ship It.Presenta: 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.
16:11ENMatthew Berman· 22 jun 2024AI News in June 2024: SSI, Tree Search, Meta, Claude and ARC PrizeMatthew Berman · Forward FutureMatthew Berman reviews new AI research, product launches and a competition, with context on historical benchmarks and uncertain claims.
16:49ENMatthew Berman· 21 jun 2024How Apple's On-Device and Private Cloud Models WorkMatthew Berman · Forward FutureMatthew Berman examines Apple's 2024 foundation-model overview, from training data and quantization to task adapters and human evaluations.
13:18ENMatthew Berman· 20 jun 2024Testing Local AI on a Dual RTX 6000 WorkstationMatthew Berman · Forward FutureMatthew Berman tests language models, concurrent inference, and image generation on a Dell Precision tower supplied by Dell and NVIDIA.
27:37ENMatthew Berman· 17 jun 2024Leo Pekelis on Extending Llama 3 to a Million-Token ContextMatthew Berman, Leo Pekelis · Forward FutureGradient’s chief scientist explains long-context training, coding use cases, inference costs and why retrieval benchmarks tell only part of the story.
1:06:04EN20VC with Harry Stebbings· 12 jun 2024Pódcast · 2 vocesAlexandr Wang on AI Data, Model Development and Building ScaleInvitado: Alexandr Wang · Presenta: Harry StebbingsScale’s founder discusses his data-bottleneck thesis, enterprise adoption, AI security and the challenges of growing a team.
1:03:12EN20VC with Harry Stebbings· 5 jun 2024Pódcast · 2 vocesAravind Srinivas on Reasoning, Model Economics and Building PerplexityInvitado: Aravind Srinivas · Presenta: Harry StebbingsThe Perplexity co-founder discusses AI scaling, foundation-model competition, advertising and the challenges of building a search product.
1:59:15ENCognitive Revolution "How AI Changes Everything"· 25 may 2024PódcastA Protein Backbone in a Minute: The Real Progress, Bottlenecks, and Risks of AI BiologyPresenta: Nathan LabenzThe White House set a thousand-fold lower reporting bar for biological models than for language models—and this conversation shows why that gap already looks prescient.
58:28ENCognitive Revolution "How AI Changes Everything"· 22 may 2024PódcastFrom Jailbreaking ChatGPT to Anthropic’s First Prompt Engineer: How Alex Albert Ships Claude 3Presenta: Nathan LabenzA year after building jailbreakchat.com, Alex Albert walks Nathan Labenz through Claude 3’s writing, orchestration, character work, and the guardrails app developers still skip.
12:50ENMatthew Berman· 22 may 2024Testing Smaug: Hosted 70B Versus a Smaller Local ModelMatthew Berman · Forward FutureMatthew Berman compares two Smaug configurations on Snake, math and reasoning puzzles.
13:37ENMatthew Berman· 16 may 2024Testing Gemini 1.5 Pro’s Long Context and Video UnderstandingMatthew Berman · Forward FutureMatthew Berman’s 2024 tests show useful image and video summaries alongside blocked code and missed details.Cargar másMostrando 24 de 381