필터 결과
38개정렬
32:54EN
Matthew Berman· 2026년 9월 10일Matthew Berman on AI Self-Improvement, Safety Warnings and His Uneasy OptimismMatthew Berman · Forward FutureA September 2026 commentary on accelerating AI research, an Anthropic resignation and the challenge of governing increasingly capable systems, with important corrections to the mathematics and benchmarks discussed.
1:53ENMatthew Berman· 2026년 7월 8일What Anthropic’s J-Space Research Suggests About AI ReasoningMatthew Berman · Forward FutureMatthew Berman discusses internal model representations, interventions and the question of consciousness, while the research itself describes important limits.
11:15ENMatthew Berman· 2026년 3월 27일ARC-AGI-3: Testing How AI Agents Learn the Rules of Unfamiliar GamesMatthew Berman · Forward FutureMatthew Berman explains ARC’s earlier grid puzzles and plays an interactive ARC-AGI-3 level, then examines early model results. The benchmark tests a specific form of adaptation rather than settling whether a system has general intelligence.
7:49ENMatthew Berman· 2025년 12월 2일DeepSeek V3.2: Reasoning, Sparse Attention and Agent TrainingMatthew Berman · Forward FutureMatthew Berman examines DeepSeek’s December 2025 model release, its benchmark results and the techniques used to improve long-context efficiency and tool use.
13:53ENMatthew Berman· 2025년 10월 10일Tiny Recursive Models and the Limits of Puzzle Benchmark ComparisonsMatthew Berman · Forward FutureMatthew Berman examines a small recursive network that reports strong ARC and Sudoku results, and what its experimental setup means.
26:09ENMatthew Berman· 2025년 7월 11일Grok 4 and Heavy Tested: Coding, Images, Reasoning and Practical QuestionsMatthew Berman · Forward FutureMatthew Berman’s July 2025 hands-on trials produce interactive prototypes and several failures, with important limits on what single examples can establish.
15:13ENMatthew Berman· 2025년 5월 10일Absolute Zero: Training a Reasoner on Self-Generated Coding TasksMatthew Berman · Forward FutureMatthew Berman examines a 2025 paper in which a pretrained language model proposes and solves code tasks using verifiable feedback.
14:04ENMatthew Berman· 2025년 4월 29일Qwen3: Open Weights, Adjustable Reasoning and Tool UseMatthew Berman · Forward FutureMatthew Berman reviews the April 2025 Qwen3 family, its benchmark claims, training pipeline and local demonstration.
19:06ENMatthew Berman· 2025년 4월 25일Sleep-Time Compute: Preparing AI Context Before a QuestionMatthew Berman · Forward FutureMatthew Berman reviews a 2025 paper on moving some reasoning ahead of a user query, and explains where its experiments found benefits and limits.
15:15ENMatthew Berman· 2025년 4월 19일Early Reactions to OpenAI o3 and o4-mini: Tool Use, Math and MistakesMatthew Berman · Forward FutureMatthew Berman reviews 2025 demos and benchmark reports, from location guessing and coding to simple failures that challenge the hype.
12:30ENMatthew Berman· 2025년 3월 7일QwQ-32B: A Smaller Reasoning Model and Its TradeoffsMatthew Berman · Forward FutureMatthew Berman examines Qwen’s reinforcement-learning recipe, tests a coding task, and compares benchmark scores and hardware requirements with DeepSeek-R1.
9:43ENMatthew Berman· 2025년 2월 20일Testing Grok 3 on Games, Reasoning, and Web SearchMatthew Berman · Forward FutureMatthew Berman tries a new set of audience-suggested prompts, from a Snake–Flappy Bird mashup to a family-logic puzzle and a tesseract visualization.
15:33ENMatthew Berman· 2025년 2월 13일Latent Reasoning and Recurrent Depth: Scaling Compute Before the Next TokenMatthew Berman, Yann Lecun · Forward FutureMatthew Berman examines a 3.5-billion-parameter research model that repeats internal computations, and asks how it relates to language, planning and chain of thought.
6:50ENMatthew Berman· 2025년 2월 12일DeepScaleR: A Small Reasoning Model Tested on Competition MathMatthew Berman · Forward FutureMatthew Berman reviews a 1.5B model’s reported benchmark results and runs a quantized version locally on one AIME problem.
12:07ENMatthew Berman· 2025년 2월 2일TinyZero: A $30 Reinforcement-Learning Experiment on a Small Arithmetic TaskMatthew Berman · Forward FutureMatthew Berman explains how a pretrained Qwen model learned to search and check answers in Countdown, and why the experiment does not recreate DeepSeek-R1.
9:58ENMatthew Berman· 2025년 2월 1일Matthew Berman Tests o3-mini on Games, Puzzles and Web SearchMatthew Berman · Forward FutureA 2025 hands-on test finds working game prototypes and several successful answers alongside a delayed Tetris row, an ambiguous puzzle and a search configuration mistake.
16:45ENMatthew Berman· 2025년 1월 27일DeepSeek-R1 and the Debate Over AI Costs and ComputeMatthew Berman · Forward FutureMatthew Berman examines the January 2025 reaction to an open-weight reasoning model, disputed cost claims and competing views of future demand for AI chips.
8:52ENMatthew Berman· 2025년 1월 21일Industry Reactions to DeepSeek R1’s 2025 ReleaseMatthew Berman · Forward FutureMatthew Berman reviews early reactions to DeepSeek R1, from open-weight reasoning models and local hardware to benchmark results and API pricing.
12:15ENMatthew Berman· 2025년 1월 20일DeepSeek-R1: Open-Weight Reasoning, Benchmark Results and a First TestMatthew Berman · Forward FutureMatthew Berman examines DeepSeek’s January 2025 release, compares reported benchmarks and prices, and tries a handful of reasoning prompts.
21:02ENMatthew Berman· 2024년 12월 24일Reactions to OpenAI’s 2024 o3 Preview: Benchmarks, Cost and LimitsMatthew Berman · Forward FutureMatthew Berman surveys the responses to the o3 preview, from mathematical benchmarks to debates about AGI, compute and reliability.
24:08ENMatthew Berman· 2024년 12월 12일What Apollo’s 2024 Scheming Tests Found in Language ModelsMatthew Berman · Forward FutureMatthew Berman reviews experiments on deceptive actions, concealment and sandbagging, with the study’s staged goals and sandbox limitations in view.
14:05ENMatthew Berman· 2024년 11월 15일Test-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.
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.
10:57ENMatthew Berman· 2024년 9월 13일Testing 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.더 보기38개 중 24개 표시