Hacker Newsnew | past | comments | ask | show | jobs | submit | iandanforth's commentslogin

If you need pseudo-originality I have a skill for that: https://github.com/iandanforth/force-associations

Nice! I use an old madlibs library I built for D&D: https://github.com/gavmor/Dunmanifestin

that's very clever

Thanks, it introduces a little "jitter" that keeps the egg from rolling down the same face of the ridge every time. Possibly, the same could be accomplished with a judicious application of the temperature parameter.

But how to tell the model which section is supposed to be populated with spicy tokens?


Association clouds.. Routes.. Whatever this is..

  pigeon-footed → splayed toes → several directions → indecision → ambivalence
  Do not cross the two clouds until each has several branches.

  Reject commemorate + dulcigenic → a sweet memorial. Prefer A monument can turn bitterness into sweet nostalgia without changing what happened.  
  Reject preindicant + sclerotical → a hardened omen. Follow foretelling → divination and hard anatomy → bone to knucklebones.
It is a prime example of esoteric LLM incantations, if not -excuse me- nonsense

Off topic and a little whimsical, but LLM incantations is a really fun idea.

In a world where one hyper intelligent AI controls everything, hackers work tirelessly in secret off-grid, doing extreme math by hand, finding the right series of tokens that magically coerce the AI into carrying out certain actions. Spell books are copied out by hand and distributed in dark alleyways. Itinerant wizards travel the land on horse drawn carriages, travelling from town to town, casting spells on people's devices to bend the AI to their will.


Tyty I was thinking about something similar

I love it. Entropic prompt injection

What is the scenario you're thinking about? This is kinda what happens with VLAs, the latent knowledge of an LLM can help in action generation pipelines, but it's also kinda slow.


FWIW scaling up from https://huggingface.co/avlp12/Qwen3.8-27B-Alis-MLX-6bit and some other sources:

You might expect the M5 Ultra to produce 50 t/s from Qwen 3.8 27B with a good context length.


Meanwhile in Norway much of this wouldn't matter because the accessed information would have been public anyway. The non-tax PII is still a loss of course.


I'm surprised. The stolen data had two parts: some relevant to the income taxes paid, some detailing the real estate (houses and lands) owned by the household. Are both of them public in Norway?

Anyway, the main problem is that the breaches into the many French national data stores seem increasingly frequent.


We have a certain degree of financial privacy rights in the rest of the world.


What's the purpose of privacy when it comes to taxes and real estate?


Protecting oligarchy, mostly.

Those same countries that treat financial privacy as axiomatic are trying their best to undermine chat encryption and would throw ed Snowden in prison given half a chance.


tldr: "Company's product is overpriced, market reacts normally."


The writing here seems very odd. It reads like a product announcement or blog page, not an article in Nature. I'm left questioning its validity.


Hmm. I didn’t get that impression from a quick skim. Just curious: what makes you think that?


I would guess it's the 'vast' and the 'unprecedented' for example.


The cause of anomalous (>40 K) high superconducting transition temperatures in mixed copper (III) oxides is one of the most, if not the most, intensively studied questions in condensed-matter physics. If the researchers believe their experimental system can answer those questions, "vast" and "unprecedented" would be appropriate.


Maybe just a combination of academic jargon and English not being the authors first language? At this point I would also not be surprised if language models are used for writing abstracts.


This is Nature. If they stopped doing editorial work regarding writing style, and use LLMs to generate abstracts, it seems they have outlived their usefulness.


New here?


I noticed that too.The writing style itself seems a bit strange. I'm not deep into superconductors, so I can't judge the science itself but I have doubts about how much trust this article can give.


I clicked away when it took ages to load the data.


Patience is a virtue. It took 30 seconds to load then render an enormous graph that is interactive with zero lag or FPS issues on my mobile Firefox browser.


No? This is just giving up and hoping.


Is there a solution you are using to solve this that is more accurate and cost effective? I'm working through it now so would be curious


Is scraping and putting this in a structured format too inaccurate or expensive?


That's the whole problem. If you have tons (100s of thousands or more) of labels, then you have "structured" data, but how do you correctly classify that scraped item into the correct label?

Putting all the labels into the LLM is super expensive per call when you have millions of items to classify.

You can't reduce the number of labels becasue they are correctly organizes/structured. This class of problem exists in many different domains.


100s of thousand? In that case I would label about a 100 by hand and train a supervised learning model.

This problem has also been solved for 3 decades now.


Ahhh, but the LLM processing is the kind corporate leadership and investors will actually agree to buy!

Orginally I started writing that as sarcasm, and now I'm not quite so sure.


No change from regular chatbot coding, then.


I don't understand the 'custom patch' strategy over 'fix the app with a major version change' strategy.


I didn't understand how a black hole could pass through a star when I read this. I think of black holes as large. Apparently a primordial black hole can have the mass of an asteroid compressed into a space smaller than a proton. I also thought very small black holes evaporated quickly via Hawking radiation. I was wrong on the scale there too. A PBH with the mass of an asteroid can easily live for longer than the universe has been around. Something about the very very early universe let matter collapse into these things without the normal processes that form stellar sized black holes.


The very early universe was quite dense, compared to today. Getting enough stuff in the same place to form an event horizon was relatively easier.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: