1
2
Surprise and Delight [video] (youtube.com)
1
Chat GPT Is Eating the World (chatgptiseatingtheworld.com)
47
Mathematics of Data Science (arxiv.org)
6
The Language of Thought Is Not Natural Language (biorxiv.org)
5
Manual Work Is a Bug (acm.org)
3
Rethinking the Evaluation of Harness Evolution for Agents (arxiv.org)
2
CSS-DOS – a '80s PC in a stylesheet (ahmedamer.co.uk)
1
A thinking thing that is no one (baronca.substack.com)
2
Mushroom trip: a mycologist's tour of the Tarkine (theguardian.com)
56
Unexpected Solidlike Fracture in Simple Liquids (quantamagazine.org)
1
A no-brainer for protecting your brain (economist.com)
4
Terrorist Groups Are Using A.I. To Gain an Edge in Battle (nytimes.com)
1
The Weird Pattern of Japan Determining the World Cup Winners (reddit.com)
1
Chatbots Can Go into a Delusional Spiral. Here’s how it happens. (nytimes.com)
3
Collapse of AMOC ocean current may be locked in (newscientist.com)
1
How to build a pyramid if Aliens won't return your calls [video] (youtube.com)
17
Billions of Sketches Reveal Hidden Cultural Variation in Human Concepts (arxiv.org)
1
LLM-as-a-Verifier: A General-Purpose Verification Framework (arxiv.org)
2
Robust Secret Storage in Networks (arxiv.org)
5
Gov. Pritzker Signs Nation-Leading Artificial Intelligence Safety Law (prezly.com)
1
Introduction to Algorithms and Machine Learning from Sorting to Strategic Agents [pdf] (justinmath.com)
8
Lord of the Roths (propublica.org)
1
Learning to Replicate Expert Judgment in Financial Tasks (thinkingmachines.ai)
24
Educators disciplined over Charlie Kirk posts are securing big payouts (nbcnews.com)
24
Return of the Nigerian Prince Redux: Beware Book Club and Book Review Scams (2025) (writerbeware.blog)
2
AI models' values are different from most people's (economist.com)
5
DOGE self-deletes on July 4th. The grand experiment fell apart long before that (eenews.net)
2
A First Course in Causal Inference (arxiv.org)
2
Relationships make us happy – and healthy (news.harvard.edu)
1
Learning to Replicate Expert Judgment in Financial Tasks (thinkingmachines.ai)
2
There's Nothing Democratic About These Socialists (theatlantic.com)
1
Baseline brain scan predicts adolescent depression and anxiety one year later (medrxiv.org)
1
Towards Automating Scientific Review with Google's Paper Assistant Tool (arxiv.org)
2
It's time to reserve your WhatsApp username (whatsapp.com)
3
There's a Reason Women Aren't Swooning over AI Like Men Are (thenoosphere.substack.com)
1
The Scaling of PEFT: Towards Million Personal Models of Trillion Parameters (arxiv.org)
8
Cheap Drones Are Rewriting Warfare (barrons.com)
1
Generative artificial intelligence creates delicious, nutritious burgers (nature.com)
1
Using Local Coding Agents (sebastianraschka.com)
4
Join .agent – Pre-Register Your Domain Free (agentcommunity.org)
1
Global warming has made Europe's heatwave 2-4°C worse (economist.com)
3
PsychoPass: Geometric Profiling of Multi-Turn Adversarial LLM Conversations (arxiv.org)
3
A Source of Mysterious Repeating Radio Signals from Space Has Been Identified (wired.com)
38
The Reversal Curse: LLMs trained on "A is B" fail to learn "B is A" (2023) (arxiv.org)
12
AI's Brokenomics (wheresyoured.at)
1
Yochai – Jewish Learning Companion (yochai.wiki)
1
The Phenomenon of Bullshit Jobs (strikemag.org)
2
Any Sufficiently Large Lookup Table Must Be Conscious (julianrdcosta.substack.com)
16
I canceled my French tutor and built an LLM tool that does it better (alshe.substack.com)
2
The No Hallucination Guarantee (hudson-labs.com)
2
We Ran the Numbers. Remote Work Is Bad for Us (nytimes.com)
6
A Trillion Dollars Isn't Worth It If You Have to Be Elon Musk (currentaffairs.org)
1
The Most Promising Ebola Vaccine Has Been Sitting on the Shelf for 15 Years (wired.com)
3
TrophyLab - Access to aggressor equipment and its analysis (gov.ua)
4
What I Learned from Being Burned Alive (wsj.com)
5
Peptides Fuel a Multibillion-Dollar Startup Race (bloomberg.com)
6
How Does One Brain Speak Two Languages? (nytimes.com)
2
How to Sparkle in Conversation with Strangers (newscientist.com)
1
Can LLM Agents Infer World Models? Evidence from Agentic Automata Learning (arxiv.org)
2
VibeThinker-3B: Exploring the Frontier of Verifiable Reasoning in Small LLMs (arxiv.org)
1
Natural Language Autoencoders Produce Explanations of LLM Activations (transformer-circuits.pub)
4
The efficiency-gain illusion: People underestimate the rate of AI use (arxiv.org)
2
Loopcraft: The Art of Stacking Loops (latent.space)
2
American Dads Rock: Fathers Are Doing More At Home Than Ever (ifstudies.org)
1
Natural Language Autoencoders Produce Explanations of LLM Activations (transformer-circuits.pub)
1
Summary of Key Changes in OMB's Proposed Federal Financial Assistance Rule (elizabethginexi.substack.com)
2
American capitalism is run by millionaires, not billionaires (economist.com)
10
The Defeat of Nuclear Deterrence (foreignaffairs.com)
1
Implicit Bias: Evolution of a Powerful Idea (annualreviews.org)
1
The social consequences of AI delegation (arxiv.org)
4
SpaceX IPO latest: Elon Musk becomes first trillionaire as opening shares soar (sky.com)
2
From Local to Global: A Graph RAG Approach to Query-Focused Summarization (arxiv.org)
5
Academic's warning about using AI was written using AI (thetimes.com)
5
The Nerdy Escorts Cashing in on Silicon Valley's AI Boom (forbes.com/sites/annatong)
6
The 90-year-old idea behind JEPA models: Canonical Correlation Analysis (shonczinner.github.io)
1
Apple Intelligence: Apple got it right this time (barrons.com)
2
The Bitter Lesson (incompleteideas.net)
51
Is grep all you need? How agent harnesses reshape agentic search (arxiv.org)
3
Sam Bankman-Fried files for a presidential pardon from Trump (theblock.co)
1
MemGraphRAG: Memory-Based Multi-Agent System for Graph RAG (arxiv.org)
101
The architecture of the internet creates risks for democracy (science.org)
17
An introduction to functional analysis for science and engineering (arxiv.org)
53
Tokenomics: Quantifying Where Tokens Are Used in Agentic Software Engineering (arxiv.org)
2
Gaia2: Benchmarking LLM Agents on Dynamic and Asynchronous Environments (arxiv.org)
1
Dallas Housing Market Cools as H-1B Visa Indian Buyers Vanish (bloomberg.com)
4
SpaceX must wait a year to be part of S&P 500 (msn.com)
1
Seven Years on a Public Clipboard: Secrets, Türkiye's Exposure, and a Stored XSS (beyondmemory.io)
66
Do transformers need three projections? Systematic study of QKV variants (arxiv.org)
1
AutoLab: Can Frontier Models Solve Long-Horizon Auto Research Engineering Tasks? (arxiv.org)
1
A Primer in Post-Training Reasoning Data: What We Know About How It Works (arxiv.org)
1
The 15-minute city is a dead end (lse.ac.uk)
2
Today a crazy quantum story just got wilder (twitter.com/drakefjustin)
4
Selective divergence between Grokipedia and Wikipedia article (pnas.org)
75
[flagged] My 15-year-old relative was killed for refusing to marry her cousin (theguardian.com)
3
Nvidia and Microsoft Reinvent Windows PCs for the Age of Personal AI (nvidia.com)
1
Emergence World: A Laboratory for Evaluating Long-Horizon Agent Autonomy (emergence.ai)
1
Who Follows Whom? (cambridge.org)
14
Britain has crushed immigration, and harmed itself (economist.com)
10