Akira Yoshiyama

on the atlas — 231

• A New Era of Midjourney10 savers
• Books1 savers
• Software Engineering in the Age of AI1 savers
• GLM-5.2: Built for Long-Horizon Tasks4 savers
• A Probabilistic Theory of Pattern Recognition1 savers
• Kearns-Vairani-1994-Introduction-to-Computational-Learning-Thoery-Ch01.pdf1 savers
• Daniel Hsu – FAQ1 savers
• [1911.08731] Distributionally Robust Neural Networks for Group Shifts: On the Importance of Regularization for Worst-Case Generalization1 savers
• [1909.02060] Distributionally Robust Language Modeling1 savers
• Hey, N00b, We Didn't Hire You to Complete Tasks2 savers
• [2506.22401] Exploration from a Primal-Dual Lens: Value-Incentivized Actor-Critic Methods for Sample-Efficient Online RL1 savers
• [2507.13181] Spectral Bellman Method: Unifying Representation and Exploration in RL1 savers
• [2507.04239] Scaling Context Requires Rethinking Attention1 savers
• [2412.00543] Evaluating the Consistency of LLM Evaluators1 savers
• The Revenge of the Data Scientist – Hamel’s Blog - Hamel Husain1 savers
• guo17a.pdf1 savers
• [2602.19362] LLMs Can Learn to Reason Via Off-Policy RL1 savers
• Future leakage in block-quantized attention | MatX1 savers
• Machine Studying | Jacob Xiaochen Li2 savers
• Adithya S K on X: "RL Coding Environments 101: Why Harbor Exists" / X1 savers
• [2603.21383] PivotRL: High Accuracy Agentic Post-Training at Low Compute Cost1 savers
• [2411.00640] Adding Error Bars to Evals: A Statistical Approach to Language Model Evaluations1 savers
• RLAlgsInMDPs.pdf1 savers
• Why I Quit Learning Languages - YouTube1 savers
• Fahim Tajwar1 savers
• Applying Statistics to LLM Evaluations1 savers
• PPO for LLMs: A Guide for Normal People1 savers
• True Agents Model the World1 savers
• Kai Goodall1 savers
• The pitfalls of hyperspecialization | Aarush Gupta1 savers
• The machines are fine. I'm worried about us.22 savers
• Nobody Cracks Open a Programming Book Anymore · unix.foo2 savers
• The Eigenprize2 savers
• principia labs5 savers
• [2604.21691] There Will Be a Scientific Theory of Deep Learning5 savers
• Drunk Post: Things I’ve Learned as a Senior Engineer1 savers
• Linoa – Tokyo Archive1 savers
• Coding Models Are Doing Too Much | wh3 savers
• Code Storage by the Pierre Computer Company2 savers
• The paper computer « the jsomers.net blog3 savers
• Quantization from the ground up | ngrok blog3 savers
• Amy Tam on X: "The Cost of Staying" / X1 savers
• <h1>Rob Pike's 5 Rules of Programming</h1>1 savers
• NL.pdf6 savers
• First, Make Me Care, by Gwern · Gwern.net5 savers
• The appropriate amount of effort is zero1 savers
• Roc Camera4 savers
• Looking for Alice - by Henrik Karlsson - Escaping Flatland84 savers
• You and Your Research76 savers
• The Bitter Lesson74 savers
• the friendship theory of everything68 savers
• Speed matters: Why working quickly is more important than it seems « the jsomers.net blog65 savers
• 2025 letter | Dan Wang62 savers
• Even if you beat me - The Dublin Review57 savers
• There’s no speed limit55 savers
• How I've run major projects | benkuhn.net50 savers
• An Opinionated Guide to ML Research50 savers
• Reality has a surprising amount of detail49 savers
• Do Thing, Do One Thing45 savers
• no good alone - by rayne fisher-quann - internet princess45 savers
• escaping flatland: career advice for CS undergrads41 savers
• To Enjoy UWaterloo39 savers
• Defeating Nondeterminism in LLM Inference - Thinking Machines Lab37 savers
• College advice for people who are exactly like me36 savers
• The Old World Is Dying: Advice for 2026 graduates36 savers
• Dario Amodei — Machines of Loving Grace34 savers
• College as an incubator of Girardian terror | Dan Wang34 savers
• Half-assing it with everything you've got30 savers
• Guide to Career Planning, part 1: Opportunity28 savers
• Research as Understanding26 savers
• Face it: you're a crazy person - by Adam Mastroianni26 savers
• Making Deep Learning Go Faster25 savers
• curius graph24 savers
• How to win a best paper award (or, an opinionated take on how to do important research)24 savers
• How To Make (almost) Anything23 savers
• Post 38: On Slack - Having room to be excited — Neel Nanda23 savers
• We Need a New Science of Progress - The Atlantic22 savers
• Research Debt22 savers
• Principles of Effective Research | Michael Nielsen20 savers
• The Last Question20 savers
• nothing, except everything. - YouTube19 savers
• Humanity's Last Machine18 savers
• Manifesto · Moth Fund18 savers
• How to Land a Frontier Lab Job17 savers
• Startup Advice - Sam Altman17 savers
• Statement from Dario Amodei on our discussions with the Department of War \ Anthropic17 savers
• flâneur — a map of Curius links17 savers
• 2025 LLM Year in Review | karpathy16 savers
• Willingness to look stupid is a genuine moat in creative work16 savers
• blogroll | benkuhn.net16 savers
• How To Understand Things - Nabeel S. Qureshi16 savers
• The Shigalyovist Turn16 savers
• EVERY SUPPLY CHAIN IN THE WORLD - by vincent huang16 savers
• What are Diffusion Models? | Lil'Log16 savers
• LOW←TECH MAGAZINE15 savers
• Deriving Muon15 savers
• 50 things I know - by Cate Hall - Useful Fictions15 savers
• The Era of Experience Paper.pdf15 savers
• Greg - Sam Altman15 savers
• advice15 savers