Résultats
232 articlesTrier
30:29EN
Matthew Berman· 3 janv. 2025How Researchers Think o1-Style Reasoning Models WorkMatthew Berman · Forward FutureMatthew Berman explains a research survey of policy initialization, rewards, search and reinforcement learning, and the open questions behind reasoning models.
1:07:32EN
Unsupervised Learning: With Jacob Effron· 18 déc. 2024Podcasto1 Is a New Generation — and the Day AGI Arrives Will Feel BoringAnime : 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.
48:18ENUnsupervised Learning: With Jacob Effron· 6 déc. 2024PodcastNoam Brown: There's No Hard Wall—Just Eight Orders of Magnitude of HeadroomAnime : Jacob EffronA ChatGPT query costs about a penny. The problems society cares about most, he says, are worth a million dollars.
27:54ENMatthew Berman· 5 déc. 2024AI News in December 2024: Genie 2, World Models, MCP and New AI ReleasesMatthew Berman · Forward FutureMatthew Berman surveys early interactive world models, voice and video tools, QwQ, distributed training and Amazon’s Nova announcements.
1:49:41EN
Cognitive Revolution "How AI Changes Everything"· 21 nov. 2024PodcastA Team of Fifteen Matched Meta’s Post-Training on Meta’s Own BaseAnime : Nathan LabenzNathan Lambert walks through Tulu 3: the three-stage recipe, the 1 percent versus 14 percent lesson, and why open post-training now lives or dies on data and evals.
14:50ENMatthew Berman· 19 nov. 2024AI News in November 2024: Mistral, Gemini, Nuclear Safety and Medical AIMatthew Berman · Forward FutureMatthew Berman surveys Le Chat updates, model rankings, nuclear safeguards and a diagnostic reasoning study.
14:05ENMatthew Berman· 15 nov. 2024Test-Time Training Improves ARC Puzzle Results in an MIT StudyMatthew Berman · Forward FutureMatthew Berman explains a 2024 paper that adapts a language model to each puzzle at inference time and combines its predictions with program synthesis.
0:35ENMatthew Berman· 6 nov. 2024An IBM Visit: Granite Models, InstructLab and Quantum ComputingMatthew Berman · Forward FutureMatthew Berman previews a longer discussion of IBM’s small AI models, model customization and quantum research.
11:56ENMatthew Berman· 3 nov. 2024IBM Granite 3.0 and InstructLab: Small Models for Enterprise AIMatthew Berman · Forward FutureMatthew Berman reviews IBM's 2024 model family and its approach to adding domain knowledge. IBM partnered on the video and invited him to its event.
2:24:18ENCognitive Revolution "How AI Changes Everything"· 30 oct. 2024PodcastMMLU Stays at Chance Until Late Training. Loss Is Only an Alarm.Anime : Nathan LabenzZyphra’s training lead walks through Zamba 2’s hybrid architecture, the evals that actually matter, and the trade secrets small labs still have to rediscover.
10:39ENMatthew Berman· 29 oct. 2024Zamba2-7B: A Hybrid Mamba–Transformer Model Tested on Coding and ReasoningMatthew Berman · Forward FutureMatthew Berman compares Zyphra’s efficiency claims with his own informal tests, including Python games, spatial puzzles and simple word tasks.
9:38ENMatthew Berman· 15 oct. 2024Liquid Foundation Models: Company Benchmarks and a Hands-On TestMatthew Berman · Forward FutureMatthew Berman reviews Liquid AI’s 2024 model and memory claims, then tries the 40B preview on coding, logic, and instruction-following prompts.
1:06:48ENUnsupervised Learning: With Jacob Effron· 3 oct. 2024Podcasto1 Is Slow and Hard to Use. Percy Liang Still Calls It a Turning PointAnime : Jacob EffronStanford’s Percy Liang splits OpenAI’s o1 into a disappointing product and the most important research signal in years.
9:15ENMatthew Berman· 25 sept. 2024Llama 3.2 Adds Vision and Smaller On-Device ModelsMatthew Berman · Forward FutureIn a Meta-partnered 2024 video, Matthew Berman reviews the model family, its tooling and a brief hosted 1B demonstration.
1:18:32ENCognitive Revolution "How AI Changes Everything"· 20 sept. 2024PodcastA DNA Model That Can Tell Which Genes You Cannot TouchAnime : Nathan LabenzStanford's Brian Hie walks through hybrid architectures, a genome-scale language model trained only on DNA, and an antibody that binds 25 times tighter than nature.
0:54ENMatthew Berman· 16 sept. 2024OpenAI’s o1 Preview: A First Look at Longer ReasoningMatthew Berman · Forward FutureMatthew Berman introduces the o1-preview and o1-mini models and tries a Python game prompt.
9:43ENMatthew Berman· 26 août 2024Matthew Berman Tests Jamba 1.5 on Coding and Reasoning PromptsMatthew Berman · Forward FutureA 2024 hands-on test of AI21’s hybrid model finds slow output in one hosted session and mixed answers to a small set of prompts.
2:04:15ENCognitive Revolution "How AI Changes Everything"· 14 août 2024PodcastA parrot cannot paraphrase — and Claude cannot be talked into harmAnime : 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.
15:24ENMatthew Berman· 13 août 2024Testing the Anonymous sus-column-r Model, Later Identified as Grok-2Matthew Berman · Forward FutureMatthew Berman tests a mysterious 2024 Arena model on coding and reasoning; its verbose answers prompted speculation about OpenAI that the later identification contradicted.
14:32ENMatthew Berman· 10 août 2024OpenAI Departures, Developer Tools and Synthetic Media: An August 2024 AI RoundupMatthew Berman · Forward FutureMatthew Berman reviews leadership changes, training-data disputes, structured API output, Qwen2-Math and emerging deepfake tools.
16:06ENMatthew Berman· 9 août 2024Q* and Strawberry: What Was Known and Speculated in August 2024Matthew Berman · Forward FutureMatthew Berman reviews contemporary rumors and research concepts around reasoning, search and self-play, while the underlying project remained undisclosed.
13:40ENMatthew Berman· 29 juil. 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 juil. 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 juil. 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.Charger plus24 sur 232 affichés