3
1
AI Consciousness, Qualia, and Personhood (dileeplearning.com)
1
Petri Dish Neural Cellular Automata (sakana.ai)
5
Transformer co-author says he's 'sick' of transformers, the tech that powers AI (venturebeat.com)
3
Sakana AI CTO says he's 'sick' of transformers that powers every major AI model (venturebeat.com)
1
ShinkaEvolve: Evolving New Algorithms with LLMs, Magnitudes More Efficiently (sakana.ai)
1
ShinkaEvolve: Evolving new algorithms with LLMs with higher sample-efficiency (sakana.ai)
1
The Science of Intelligent Exploration: Why We Need Exploration in AI (richardcsuwandi.github.io)
10
Gamblers Now Bet on AI Models Like Racehorses (wsj.com)
1
ICML Statement about subversive hidden LLM prompts (icml.cc)
101
Evaluating publicly available LLMs on IMO 2025 (matharena.ai)
1
AI Is Evolving – and Changing Our Understanding of Intelligence (noemamag.com)
1
AI Wreaked Havoc on the Lo-Fi Beat Scene (pitchfork.com)
1
Inference-Time Scaling and Collective Intelligence for Frontier AI (sakana.ai)
3
Zero Zen (nymag.com)
1
Zero Zen: Cheap yen and ChatGPT turned Kyoto into the loveliest tourist trap (nymag.com)
2
Inference-Time Scaling and Collective Intelligence for Frontier AI (sakana.ai)
2
Reinforcement Learning Teachers of Test Time Scaling (sakana.ai)
1
Towards Automating Long-Horizon Algorithm Engineering for NP-Hard Problems (sakana.ai)
38
How to Build Conscious Machines (osf.io)
6
Jensen Huang says he disagrees with almost everything Anthropic CEO Amodei says (fortune.com)
2
Text-to-LoRA: Instant Transformer Adaption (arxiv.org)
1
Darwin Gödel Machine: A Commentary on Novelty and Implications (4m4.it)
53
A deep dive into self-improving AI and the Darwin-Gödel Machine (richardcsuwandi.github.io)
5
Darwin Godel Machine: Open-Ended Evolution of Self-Improving Agents (arxiv.org)
2
The Darwin Gödel Machine: AI that improves itself by rewriting its own code (sakana.ai)
1
Sudoku-Bench Leaderboard (sakana.ai)
75
Continuous Thought Machines (sakana.ai)
1
Sakana AI super-powers AI reasoning using Japan's own Sudoku Puzzles (sakana.ai)
1
The AI Scientist Generates Its First Peer-Reviewed Scientific Publication (sakana.ai)
1
Show HN: AI Scientist Generates Its First Peer-Reviewed Scientific Publication (sakana.ai)
1
The AI Scientist Generates Its First Peer-Reviewed Scientific Publication (sakana.ai)
4
The AI CUDA Engineer: Agentic CUDA Kernel Discovery, Optimization, Composition (sakana.ai)
2
Evolving Deeper LLM Thinking (arxiv.org)
39
Transformer^2: Self-Adaptive LLMs (sakana.ai)
40
Automating the search for artificial life with foundation models (sakana.ai)
2
I sensed anxiety and frustration at NeurIPS'24 (kyunghyuncho.me)
1
Population-Based Model Merging via Quality Diversity (sakana.ai)
1
Russia fines Google more than entire GDP for blocking YouTube accounts (sky.com)
2
Japan is remarkably open to AI, but slow to make use of it (economist.com)
2
Intelligence at the Edge of Chaos (arxiv.org)
1
Can LLMs Generate Novel Research Ideas? A Large-Scale Human Study (arxiv.org)
6
Is Xi Jinping an AI Doomer? (economist.com)
10
Workers at Google DeepMind Push Company to Drop Military Contracts (time.com)
2
Automated Design of Agentic Systems (arxiv.org)
3
JPEG-LM: LLMs as Image Generators with Canonical Codec Representations (arxiv.org)
69
The AI Scientist: Towards Automated Open-Ended Scientific Discovery (sakana.ai)
3
The 'godmother of AI' has a new startup already worth $1B (theverge.com)
1
The Tragedies of Reality Are Coming for You (alexirpan.com)
1
Artificial General Intelligence (mitpress.mit.edu)
165
Language is primarily a tool for communication rather than thought [pdf] (gwern.net)
1
AlphaMath Almost Zero: process Supervision without process (arxiv.org)
100
Creativity has left the chat: The price of debiasing language models (arxiv.org)
25
Can LLMs invent better ways to train LLMs? (sakana.ai)
2
Kan: Kolmogorov-Arnold Networks (arxiv.org)
1
Why the Military Can't Trust AI (foreignaffairs.com)
3
Uncompetitive Programming (Part 1) (sujitkc-techbits.blogspot.com)
1
Securing Canada's AI Advantage (pm.gc.ca)
1
Collective intelligence: A unifying concept integrating biology across scales (nature.com)
2
Evolutionary Optimization of Model Merging Recipes (sakana.ai)
2
Evolving New Foundation Models: Unleashing Power of Automating Model Development (sakana.ai)
1
Evolving New Foundation Models: Unleashing Power of Automating Model Development (sakana.ai)
2
Evolving New Foundation Models: Unleashing Power of Automating Model Development (sakana.ai)
1
The Ethos of the Divine Age (piratewires.com)
2
Are polynomial features the root of all evil? A myth (alexshtf.github.io)
2
Everyone Could Have Died in the Tokyo Airport Crash. Here's Why They Didn't (nymag.com)
1
The roboticist who wants to bring AI into contact with the real world (newscientist.com)
99
Fortran 2023 (iso.org)
1
Fortran 2023 has been published (iso.org)
2
Stability AI Gets Intel Backing in New Financing (bloomberg.com)
5
Design giant Ideo cuts 1/3 staff, close offices as era of "Design Thinking" ends (fastcompany.com)
2
The Techno-Optimist Manifesto (twitter.com/pmarca)
1
Where AI Meets Copyright Law, Present and Future (medium.com/vkhosla)
1
Strategic Game Datasets for Enhancing AI Planning: Invitation for Collaboration (laion.ai)
41
Be an Open Source Absolutist (twitter.com/id_aa_carmack)
106
[dupe] John Carmack and Rich Sutton partner to accelerate development of AGI (amii.ca)
2
The Alberta Plan for AI Research (arxiv.org)
5
Teens Love LinkedIn (thecut.com)
3
Twitter stops showing headlines because Elon thinks it makes posts look better (theverge.com)
17
Self-Assembling Artificial Neural Networks Through Neural Developmental Programs (arxiv.org)
1
What is at stake in the AI based techno-economic war between the West and China? (medium.com/vkhosla)
2
‘Equity-Based Weeding Process’ Clears Ontario Public School Bookshelves (nationalreview.com)
2
The USAF Pairs Piloted Jets with AI Drones (ieee.org)
1
Generative AI and Intellectual Property (ben-evans.com)
2
Generative AI and Intellectual Property (ben-evans.com)
1
Why scientists are delving into the virtual world (nature.com)
0
Analogies for thinking about intelligence as a quantity (togelius.blogspot.com)
2
To Compress or Not to Compress- Self-Supervised Learning and Information Theory (arxiv.org)
63
The Albanian town that TikTok emptied (codastory.com)
49
MatX: Faster Chips for LLMs (matx.com)
3
AI Startup Hugging Face Is Raising Fresh VC Funds at $4B Valuation (forbes.com/sites/alexkonrad)
3
HyperDreamBooth: HyperNetworks for Fast Personalization of Text-to-Image Models (hyperdreambooth.github.io)
1
Stable Doodle (clipdrop.co)
1
Eight AI Startups Winning the Race for Tech Talent (medium.com/lightspeed-venture-partners)
1
Show HN: Stable Diffusion Reimagine XL (clipdrop.co)
1
Jürgen Schmidhuber, ‘Father Of Modern AI’, says AI won’t lead to dystopia (reddit.com)
5
Bringing Open Large Language Models to Consumer Devices (mlc.ai)
78
Advocating for Open Models in AI Oversight: Stability AI's Letter to U.S. Senate (stability.ai)
1
Self-Alignment of Language Models from Scratch with Minimal Human Supervision (arxiv.org)
3