Jacob Tan En
@jacob_tan_en
Maths, stats, econs / software, functional programming / ja-JP, zh-CN
Bạn có thể thích
To search for a reason (i.e. cause) for all existence, is equivalent to looking for an object (i.e. explanation) outside of the universal set. It is a futile endeavour. By definition, there exists no object outside the universal set.
5 years later and I’ve made no further progress on this one: waitbutwhy.com/table/why-is-t…
Bret Weinstein just said something that won’t leave my head: For the first time in 300,000 years of human evolution, we removed the cost from the single biggest reward nature ever invented — sex and pair-bonding. Reliable birth control + abortion = you can now cash the…
ur procrastinating because ur thinking of what u SHOULD do but not thinking about what u CAN do. you're operating on hopes and dreams so ofc ur paralyzed
“There’s no way metabolism is involved in so many diseases.” — a confident declaration from someone who’s never paused to consider how cells actually work.
Large Language Models, in their current form, are a big nothing burger. Some people are excited about the nothing burger. Some people are terrified of the nothing burger. Silicon Valley is breathlessly slavering over new ways they can sell the nothing burger. Our…
AI DEFENDING THE STATUS QUO! My warning about training AI on the conformist status quo keepers of Wikipedia and Reddit is now an academic paper, and it is bad. — Exposed: Deep Structural Flaws in Large Language Models: The Discovery of the False-Correction Loop and the…
“A guy could starve to death in this town, there are so many restaurant choices.” Decision fatigue
"robot also lacks human commonsense" Lack of commonsense is a fundamental limitation of current AI tech. Everyday human interaction leaves much unsaid, relying on implicit knowledge already shared by the participants. That which is unspoken cannot be perceived by the AI.
Dr Fei-Fei-Li explains with a simple example how everyday household chores are so extremely difficult for Robots. "If you tell a robot to open the top drawer and watch out for the vase, this is actually a really hard task for robots." because the robot must ground language into…
Summary: subsidizing demand simply raises prices without raising overall consumption, when the supply is price inelastic e.g. due to limits imposed by the same government giving out the subsidies.
Buyers: "How much will this house cost me?" Sellers: "What's your budget?" Buyers: "Well, it was 500K, but with these new fifty year mortgages, I think it could stretch to million." Sellers: "I have an astonishing coincidence to report." Look, I don't know exactly who's…
One of my favourite Jungian quotes: “The foundation of all mental illness is the unwillingness to experience legitimate suffering.”
The last few months have been devastating for LLM dreams: • Apple reasoning paper and the ASU mirage paper and many others confirmed that LLMs still can’t solve distribution shift. • GPT-5 came late and fell short. • Turing Award winner Rich Sutton thanked me for my…
At first, I thought GPT-5 had cracked those math problems on its own. Turns out (as Demis pointed out) GPT-5 just looked up the answers via web search. We really need better peer review for these “AI discovers science/math” claims.
The reason it is so important for everyone to keep pretending that AGI is definitely right around the corner is that there is now over $1T of investment riding on this belief (either already expended, or committed) Current (and recent past) capex cannot be justified by current…
So much corporate red tape can be traced to a single anecdote where someone fucked up and their manager didn't have the balls to tell them directly. So they made a policy instead. Don't be that coward. Just tell Ben.
america is betting its entire economy on an overhyped, unreliable slop-generating technology that is unproven and may take many years to pay off, if it ever does. don’t be surprised if that turns out badly.
It's still baffling to me that OpenAI thought they could build a physics engine simply by fitting a model on next frame prediction. And why are so called frontier artificial intelligence (AI) labs only knowing about this limitation now?
Fei-Fei Li says language models are extremely limited. This @GoogleDeepMind paper makes almost the same point, just in the world of video. The models are just very advanced pattern matchers. They can recreate what looks like reality because they’ve seen so much data, but they…
GenAI models “often match patterns instead of truly reasoning” Say it to yourself over and over til you full understand it. The amount of confirmation that is coming this year for my basic view is insane.
🧬 Bad news for medical LLMs. This paper finds that top medical AI models often match patterns instead of truly reasoning. Small wording tweaks cut accuracy by up to 38% on validated questions. The team took 100 MedQA questions, replaced the correct choice with None of the…
We don't train AI on data from the world. We train it on words people have said about the world. A stream doesn't rise higher than its source.
The reason you struggle to fall asleep at night is because your brain won't shut off. In the past, during moments of boredom or downtime, your mind had the space to process thoughts, prepare for stressful situations, and remind you of things you needed to do. But now, you fill…
I didn't go to college until I was 30. This gave me a chance to see it with the perspective of an adult. One lecture in Industrial Psychology, in particular, I will never forget. The professor spoke about how an effective job description focused on concretely measurable…
Seeing so many men happy to replace us with robots is very blackpilling ngl
A basic way to avoid bad habits is to imagine how you’ll feel afterward. It’s called episodic future thinking. Previewing the unwanted consequences of procrastinating, overeating, and doomscrolling can motivate better choices. Your future self is wiser than your present self.
Terence Tao reporting on the same problem we have seen endlessly in other domains: LLM’s produce output that “looks” correct but is often deeply wrong and even stupid on careful inspection. I know of no domain in which this is NOT the case. And yet people seem surprised over…
Terence Tao says today's AIs pass the eye test -- but fail miserably on the smell test. They generate proofs that look flawless. But the mistakes are subtle, and strangely inhuman. “There's a metaphorical mathematical smell.. it's not clear how to get AI to duplicate that.”…
You might feel like you're running out of time, but the real issue is likely mental energy. You sit down to work, but starting feels impossible. After an hour or two, you are emotionally drained, so you escape to YouTube for funny videos. Refreshed but guilty, you return to…
United States Xu hướng
- 1. Cowboys 73.3K posts
- 2. #heatedrivalry 24.4K posts
- 3. LeBron 105K posts
- 4. Gibbs 20.2K posts
- 5. Lions 91K posts
- 6. Pickens 14.5K posts
- 7. scott hunter 5,149 posts
- 8. fnaf 2 25.9K posts
- 9. #OnePride 10.5K posts
- 10. Paramount 19.9K posts
- 11. Shang Tsung 29.2K posts
- 12. Warner Bros 20.9K posts
- 13. Ferguson 10.8K posts
- 14. Brandon Aubrey 7,370 posts
- 15. Eberflus 2,676 posts
- 16. Ceedee 10.6K posts
- 17. #PowerForce N/A
- 18. #criticalrolespoilers 2,062 posts
- 19. Cary 38.5K posts
- 20. Goff 8,701 posts
Something went wrong.
Something went wrong.