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

Yes, I havent seen research on this tactic specifically for language models but for generative image models you can find research thatindicates you get far more variety and diversity in the images returned when suffixing the context with "noisy context" such as random letters or something else unrelated that wont "confuse" it or make it go off the rails

Found this interesting paper today while searching around the keywords "attractor language networks". This part I found especially interesting:

*Attractor Models outperform existing models across two regimes, large-scale language-model pretraining and reasoning with tiny models.*


Im usually on Firefox and I haven't run into this. But I also spent the past 2 days building a Knitsail VM solver in Rust for pulling Google Search results. Works nicely, I gave chatgpt pro (6) a minified JS bundle, and some variables and entries into the VM and told it to reconstruct the VM in Rust. Only took a couple hours to get a zip with a solver I could plug into my crawler, made a few mistakes but codex fixed those up with minimal guidance. No more Google captchas for me :)

I unfortunately live in Alberta and I have to say Smith seems decently smart. Which is not to say I agree with or would vote for her but that shes very capable of getting things she wants. Seeing Alberta pop up on HN so often is disheartening. I hope she gets voted out soon or I just might have to move

Right, I submitted because in addition to being from Alberta (originally, not there now) and having my whole family out there I think there's actually a useful analysis where Alberta's right wing forces have been for sometime a "vanguard" the further side of conservative ideology in North America ... and have had enormous "success" in not just fighting climate change initiatives but breaking apart a lot of the Canadian social contract. They're successful in enacting policy that even their own domestic population is wildly opposed to. Over and over again. E.g. they have extremely little support within their own urban centres, but still run the province however they want.

So actually, it's a place that needs to be watched as a barometer of the future, really. And an example of why "technology will save us from climate change" is a bad take: these people (petroleum sector) will fight tooth and nail to hold onto what they have and they will take us all down with them.


Lol Alberta absolutely mogs Canada in nearly every objective measurement.

Highest per capita GDP, most affordable housing for places you'd actually want to live (Calgary twice as affordable as Toronto or Vancouver), most private sector jobs, lowest taxes, etc...

It's why Alberta absolutely dominates inter-provincial migration. Thinking Alberta isn't the best province or that there's something wrong with the government has no basis in objective reality.

You should move to another Canadian province for a year, just to see.

I mistakingly moved to BC because my wife was sick of the cold and it's basically 3rd world compared to Alberta (apart from a small slice of Vancouver which is nice I guess).


It just has oil. While it is nice having low taxes because of all the oil money, it can also be draining being surrounded by people who want to be annexed by Trump, or hold socially regressive views. But economically yes it's a great province.

As opposed to the very productive and diversified Canadian economy? Over 50% of Canada's exports are natural resources and the rest of the economy is basically just downstream from that.

Where are the tech giants or manufacturing giants? We have 1 tech giant who constantly criticizes Canada, 1 large manufacturer that gets bailed out twice a decade and otherwise we're just a country of housing schemes, financial services and fast food/retail.


It's funny the housing angle that parent poster is playing, too. While Edmonton at least has made sensible decisions around its housing stock, the real reason why Alberta has "affordable housing" is that the absolutely brutally cyclical and volatile nature of the oil industry there means that every few years a wave of unemployment and poor economics knocks real estate prices down. Every time there's low oil prices there's hand wringing all over Alberta. It's just a volatile place to invest.

Alberta real estate is affordable relative to the rest of Canada because it's actually not a good investment. Not that I think real estate should be treated as an investment, and this isn't a knock per se, but it's also nothing to brag about that every few years scores of people in Alberta are losing their shirts on real estate.

I'll probably buy a house next year in Edmonton for my daughter to use (and rent out) while she's at the UofA, but I fully expect to either lose money or not gain on it.


I think it is more that land is comparatively cheap. Edmonton and Calgary are incredibly sprawly cities, with effectively infinite room to grow.

Contrast with Vancouver, which is hemmed in by ocean, mountains, and a border.


Strangely, actual empty rural land in Alberta seems remarkably expensive per acre compared to Ontario (where I own 6 acres).

Can't comment without knowing more specifically what you're comparing. Massive variation across both provinces, of course.

The standard unit in rural Alberta is the quarter section - 160 acres (half mile by half mile). Greater Edmonton is surrounded by mostly flat quarters in every direction as far as the eye can see.


Yes, my parents own a quarter section near Evansburg

VLM != LLM. Vision language models basically treat text tokens and image tokens the same. Post-training an LLM on images+text can improve its capabilities. Id recommend searching around the keyword VLM to find more resources on how multi-modal AI works.

- https://huggingface.co/blog/vlms

- https://en.wikipedia.org/wiki/Multimodal_learning


This is a fun rabbit hole I discovered. I'm currently reading the 2026 PDF: https://sigbovik.org/2026/proceedings.pdf which covers many topics that will likely pique the interest of the crowd here

I mean you might be able to if you squint hard enough but the experience of using it would be absolitely piss poor so no. Realistically speaking AI is not even nearly that good yet and buildong a browser is an absolutely huge undertaking

The sci-fi hand wringing is embarrassing. A huge chunk of the population is afflicted with mental health issues and the best psychitrists can seem to do is "Oh bro, wanna try some ketamine? Anti-psychotics?? Maybe some SSRIs, how about psychdelics?!" - this is a joke. Im all for more neuroscience research in every form because our contemporary understanding is embarassingly poor. This is a research tool for understanding what visual brain regions encode, not a mind-control device. Better models of brain function may eventually improve diagnosis and treatment of mental illness. Calling every neuroscience advance "Black Mirror" is not ethical insight. It saddens me the reaction so many commentors have here, it is just lazy technological illiteracy imo.


In a society that wasn't visibly hell-bent on monetizing every discovery in a cartoonishly evil way, I'd probably agree with you.


I personally just use pipes and tesseract, no need for another tool imo:

flameshot gui --raw | tesseract stdin stdout -l eng | xclip -in -selection clipboard



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

Search: