Résultats
276 articlesTrier
10:57EN
Matthew Berman· 13 sept. 2024Testing OpenAI o1-preview: Tetris, Physical Reasoning and a North Pole ErrorMatthew Berman · Forward FutureIn September 2024, Matthew Berman runs the new reasoning-model preview through a personal rubric and examines where its answers succeed or fail.
1:31:56ENMatthew Berman· 12 sept. 2024Testing OpenAI o1 on Puzzles, Code and Reasoning: A Launch-Day LivestreamMatthew Berman · Forward FutureMatthew Berman tries o1-preview and o1-mini on word counts, Tetris, physical reasoning and a word puzzle, with viewers checking the results.
19:38ENMatthew Berman· 9 sept. 2024Reflection 70B: Conflicting Results and Unanswered QuestionsMatthew Berman · Forward FutureMatthew Berman reviews his early coverage, independent evaluations and competing claims about the September 2024 release.
12:03ENMatthew Berman· 6 sept. 2024Reflection 70B: Self-Correction Claims and the Limits of a 2024 DemoMatthew Berman · Forward FutureMatthew Berman reviews a model trained to generate reflection steps. Its initial benchmark claims later proved difficult for independent evaluators to reproduce.
35:30ENMatthew Berman· 6 sept. 2024Reflection 70B: Its Creators Explain the Proposed Self-Correction MethodMatthew Berman, Matt Shumer, Sahil Chaudhary · Forward FutureA launch-day discussion with Matt Shumer and Sahil Chaudhary covers synthetic data, selective reflection and evaluation limits.
1:54:17EN
Cognitive Revolution "How AI Changes Everything"· 28 août 2024PodcastMetaculus Flagged Covid in January 2020. Now It Is Putting AI on the Forecasting DeskAnime : Nathan LabenzNathan Labenz talks with Metaculus CEO Deger Turan about epistemic security, a year-long bot tournament, and why a useful forecast is harder than an accurate one.
1:12:31EN
Sequoia Capital· 27 août 2024PodcastA Bare Model Closes 35% of Airline Changes. Sierra's Agents Clear 70% of Real Tickets.Clay Bavor on why putting a raw LLM in front of customers does not math, and how AgentOS, supervisor agents, and outcome pricing change the arithmetic.
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.
11:39ENMatthew Berman· 15 août 2024Grok 2 Beta: Matthew Berman Tests Reasoning and FLUX Image GenerationMatthew Berman · Forward FutureA 2024 look at xAI’s Grok 2 and Grok 2 mini covers early benchmark claims, a failed reasoning test and image-generation experiments.
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.
57:55ENMatthew Berman· 12 août 2024A First AI Livestream: Model Rumors, RAG, Agents, and Audience QuestionsMatthew Berman · Forward FutureMatthew Berman tests a livestream in August 2024 and answers viewers’ questions about AI tools, model evaluation, company knowledge, and future agents.
9:51ENMatthew Berman· 7 août 2024Strawberry Rumors and Two Anonymous Chatbot Reasoning TestsMatthew Berman · Forward FutureIn August 2024, Matthew Berman examined speculation about OpenAI’s next model and tested mystery chatbots on simple reasoning puzzles. Later disclosures clarified the identity of one of them.
12:17ENMatthew Berman· 7 août 2024Testing Gemini 1.5 Pro on Code, Reasoning, Images and VideoMatthew Berman · Forward FutureMatthew Berman compares standard and experimental Gemini versions in an informal 2024 review, finding mixed results and repeated output errors.
11:34ENMatthew Berman· 1 août 2024Testing Mistral Large 2: Coding, Reasoning and Safety in 2024Matthew Berman · Forward FutureMatthew Berman tests Mistral’s 123-billion-parameter model with small coding tasks, logic puzzles and ethical hypotheticals. The results are promising but narrowly scoped.
42:08ENSequoia Capital· 30 juil. 2024PodcastMeta Trained a 405B Teacher on 16,000 GPUs — Then Told the World to Distill ItTwo days after Llama 3.1 405B shipped, Meta's generative-AI product lead walks Sequoia through why the most expensive model in the family is free, and why he thinks the frontier itself is heading toward commodity.
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.
5:50ENMatthew Berman· 27 juil. 2024A Mixture-of-Agents Demo Using Groq and StreamlitMatthew Berman · Forward FutureMatthew Berman walks through a 2024 Groq-powered multi-agent project, from setup to inspecting its layered responses.
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.
13:19ENMatthew Berman· 20 juil. 2024Testing GPT-4o Mini Against GPT-4o at LaunchMatthew Berman · Forward FutureMatthew Berman compares OpenAI's 2024 small model on coding, reasoning and image prompts, with a sponsored HP and Intel laptop segment.
8:52ENMatthew Berman· 14 juil. 2024Yi-Large Tested: Coding, Arithmetic and Logic PuzzlesMatthew Berman · Forward FutureMatthew Berman puts 01.AI's hosted model through a small set of coding and reasoning prompts, with mixed results and some questionable grading.
25:56EN
No Priors: AI, Machine Learning, Tech, & Startups· 11 juil. 2024Podcast · 2 voixA Prompt Cannot Hold a Good Design: Six AI Practitioners at Midyear 2024Anime : Sarah Guo, Elad GilSarah Guo and Elad Gil stitch together clips from Stripe, Figma, Figure, OpenAI's Sora team, Cognition, and Scale into a half-year snapshot of what people actually shipping AI were willing to claim.
1:16:22ENCognitive Revolution "How AI Changes Everything"· 9 juil. 2024PodcastHe Spent 20 Hours Crafting a Prompt. DSPy Beat Him AnywayAnime : Nathan LabenzSander Schulhoff walks through The Prompt Report — a 78-page survey of what actually works when you prompt large language models.
8:53ENMatthew Berman· 8 juil. 2024RouteLLM: Routing Queries Between Strong and Cheaper ModelsMatthew Berman · Forward FutureMatthew Berman explains a 2024 research framework that uses preference data to choose a model for each query.
59:11ENSequoia Capital· 25 juin 2024PodcastDon't Train a Model. Factory's Founders Say Go Dig for GoldMatan Grinberg and Eno Reyes tell Sequoia's Training Data why they sell a fleet of job-specific droids on top of other people's models — and how that stack just took SWE-bench from 14% to 19%.Charger plus24 sur 276 affichés