필터 결과
136개정렬
9:00EN
Matthew Berman· 2024년 12월 20일Best-of-N Jailbreaking: What Repeated Prompt Variations Revealed in 2024Matthew Berman · Forward FutureMatthew Berman reviews research on how repeated input variations can expose safety failures in text, vision and audio models.
1:49:41EN
Cognitive Revolution "How AI Changes Everything"· 2024년 11월 21일팟캐스트A Team of Fifteen Matched Meta’s Post-Training on Meta’s Own Base진행자 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· 2024년 11월 19일AI 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.
1:23:37ENCognitive Revolution "How AI Changes Everything"· 2024년 10월 23일팟캐스트Nine of the top ten research ideas were Claude's진행자 Nathan LabenzStanford PhD student Chenglei Si ran a blind study pitting Claude 3.5 Sonnet against more than a hundred NLP researchers on novel research ideas.
2:37:47ENCognitive Revolution "How AI Changes Everything"· 2024년 10월 19일팟캐스트The intern who wrote GELU now wants humanity's last exam진행자 Nathan LabenzDan Hendrycks on activation functions, five-year benchmarks, jailbreaks that do not yield to scale, and stacking defenses because no single layer is trustworthy.
2:25:53ENCognitive Revolution "How AI Changes Everything"· 2024년 10월 16일팟캐스트When GPT-4's Base Model Introduced Itself, 75% of the Answers Were About AI진행자 Nathan LabenzOwain Evans tells Michael Trazzi why situational awareness and out-of-context reasoning are leading indicators of deceptive alignment, and why a pretrained model with almost no self-knowledge still kept talking about being an AI.
26:43ENMatthew Berman· 2024년 10월 10일Testing OpenAI o1 on Planning, Rules and Spatial ReasoningMatthew Berman · Forward FutureMatthew Berman reviews a 2024 pilot study comparing o1-preview, o1-mini and GPT-4 on six planning domains.
1:09:09ENCognitive Revolution "How AI Changes Everything"· 2024년 10월 2일팟캐스트30 Percent on SWE-bench: Cosine Taught Genie How Engineers Actually Think진행자 Nathan LabenzAlistair Pullen tells Latent Space how Cosine reconstructed human problem-solving, fine-tuned on billions of tokens, and took the top score on software engineering’s hardest public test.
11:14ENMatthew Berman· 2024년 9월 18일Pixtral 12B: Testing a 2024 Vision-Language ModelMatthew Berman · Forward FutureMatthew Berman runs informal image, text and interface-generation tests of Mistral’s Pixtral on a sponsor-provided cloud GPU.
18:09ENMatthew Berman· 2024년 9월 17일AI News in September 2024: Copilot, NotebookLM, o1 and World ModelsMatthew Berman · Forward FutureMatthew Berman reviews a week of AI announcements, from workplace assistants and audio overviews to benchmark results, generative video and spatial AI.
21:21ENMatthew Berman· 2024년 9월 12일OpenAI o1’s 2024 Launch: Reasoning Benchmarks and Early TestsMatthew Berman · Forward FutureMatthew Berman reviews o1-preview and o1-mini, OpenAI’s reported evaluations, and his own early coding experiments.
19:38ENMatthew Berman· 2024년 9월 9일Reflection 70B: Conflicting Results and Unanswered QuestionsMatthew Berman · Forward FutureMatthew Berman reviews his early coverage, independent evaluations and competing claims about the September 2024 release.
1:54:17ENCognitive Revolution "How AI Changes Everything"· 2024년 8월 28일팟캐스트Metaculus Flagged Covid in January 2020. Now It Is Putting AI on the Forecasting Desk진행자 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.
50:20EN
20VC with Harry Stebbings· 2024년 8월 28일팟캐스트 · 2명Arvind Narayanan on AI Scaling, Real-World Risks and the Limits of Model Benchmarks게스트 Arvind Narayanan · 진행자 Harry StebbingsThe Princeton professor challenges simple claims about scaling, AGI timelines and AI adoption while discussing deepfakes, healthcare, education and security.
18:39ENMatthew Berman· 2024년 8월 17일AI Scientist, Grok-2 and Agent Q: Matthew Berman’s August 2024 AI News RoundupMatthew Berman · Forward FutureA week of model releases, web agents and AI research raises questions about benchmarks, automation and the gap between prototypes and predictions.
9:58ENMatthew Berman· 2024년 8월 2일Abstract AI: A Proposed Routing Layer for Language ModelsMatthew Berman · Forward FutureMatthew Berman outlines a startup idea for choosing models and workflows by cost, speed and quality.
15:50ENMatthew Berman· 2024년 7월 23일Llama 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.
1:56:03ENCognitive Revolution "How AI Changes Everything"· 2024년 7월 13일팟캐스트Humans Near a Perfect Score. The Best AI Still Gets 39%진행자 Nathan LabenzZapier co-founder Mike Knoop and François Chollet put $1 million on ARC-AGI, a 2019 test of sample-efficient intelligence that language models still cannot crack under a one-cent-per-puzzle budget.
55:13EN
Sequoia Capital· 2024년 7월 2일팟캐스트The Best AI Scores 39% on This Test. Zapier’s Co-Founder Put $1 Million on Beating ItFrançois Chollet designed ARC to measure skill-acquisition efficiency; five years later humans still crush it and frontier models do not.
10:57ENMatthew Berman· 2024년 6월 24일Testing Claude 3.5 Sonnet: Coding, Reasoning, and VisionMatthew Berman · Forward FutureMatthew Berman puts Anthropic’s June 2024 model through informal coding, math, puzzle, and image-understanding prompts.
16:11ENMatthew Berman· 2024년 6월 22일AI 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.
26:00EN
No Priors: AI, Machine Learning, Tech, & Startups· 2024년 6월 11일팟캐스트 · 2명Why a Zapier founder thinks AGI research has been stalled for four years진행자 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.
17:03ENMatthew Berman· 2024년 6월 8일Qwen2 0.5B and 72B: A Hands-On ComparisonMatthew Berman · Forward FutureMatthew Berman tests coding, arithmetic and reasoning prompts on the smallest and largest Qwen2 models in a 2024 demonstration.
14:25ENMatthew Berman· 2024년 6월 2일Testing Phi-3 Vision, LLaVA-Llama 3 and GPT-4o on ImagesMatthew Berman · Forward FutureMatthew Berman compares three 2024 vision setups on image descriptions, screenshots, a meme and table extraction, with mixed results.더 보기136개 중 24개 표시