frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

AnkiDroid: Google Play no longer allowing Open Collective donation link

https://github.com/ankidroid/Anki-Android/issues/21656
150•hexa555•1h ago•13 comments

44% on ARC-AGI-1 in 67 cents

https://mvakde.github.io/blog/44-on-arc-1/
67•porridgeraisin•1h ago•18 comments

Fastpotify

https://fastpotify.rocks/
469•nreece•8h ago•251 comments

GPU World

https://www.gpuworld.org/
270•simonpure•8h ago•144 comments

American Airlines' Legendary Mechanic Passes Away at 100 After 80-Year Career

https://simpleflying.com/american-airlines-mechanic-passes-away-100-record-80-years/
80•NaOH•2d ago•24 comments

Tmp.0ut Volume 5

https://tmpout.sh/5/
91•ghuntley•12h ago•24 comments

Restroom Archive

https://restroomarchive.com
155•jcalx•8h ago•38 comments

Polishing Cloth Updated

https://512pixels.net/2026/08/polishing-cloth-updated/
42•zdw•6d ago•33 comments

Playa Phone

https://playaphone.com/
651•cutoff•20h ago•212 comments

I turned my security cameras into an automatic bird identification system

https://jasontucker.blog/how-i-turned-my-security-cameras-into-an-automatic-bird-identification-s...
521•speckx•18h ago•127 comments

Terence Tao explains 6 essential mathematical concepts [video]

https://www.youtube.com/watch?v=OOMx2BHHWtE
451•matthewsinclair•1d ago•63 comments

A walkable ASCII cyberpunk city in one HTML file [video]

https://www.youtube.com/watch?v=3YtygAx_C6A
339•keithcarolus•17h ago•45 comments

RotaryCell: Making an unmodified rotary phone work over LTE with an ESP32-S3

https://github.com/fregacmols/RotaryCell
78•jombib•9h ago•10 comments

CollectWise (YC F24) Is Hiring

https://www.ycombinator.com/companies/collectwise/jobs/7U500pP-founding-customer-success-engineer
1•OBrien_1107•4h ago

Dwarf Fortress is getting the mother of all magic updates

https://www.rockpapershotgun.com/dwarf-fortress-is-getting-the-mother-of-all-magic-updates-extend...
429•Tomte•4d ago•176 comments

Apple caught off guard by AI demand for Mac Mini and Mac Studio

https://www.macrumors.com/2026/08/30/apple-unexpected-mac-mini-and-studio-demand/
422•thm•22h ago•476 comments

NASA: Fill in a name, and you can have an exclusive coordinate in the universe

https://science.nasa.gov/mission/roman-space-telescope/adopt-a-pixel/
17•BearBest•1h ago•18 comments

Evidence of Fraud in an Influential Study About Procrastination

https://datacolada.org/138
259•Anon84•11h ago•178 comments

Show HN: Laser Graffiti

https://laser.consti.de
200•con•2d ago•42 comments

I think the military commissary's freezers were hacked

https://signalandsilence.substack.com/p/i-think-someone-hacked-the-commissary
362•jcurbo•23h ago•200 comments

Cheap GPS jammers are filling the world with navigation dead zones

https://www.wsj.com/tech/gps-jammers-dead-zones-e76f3261
153•vinnyglennon•1d ago•146 comments

Codes – typed application outcomes for Kotlin and Java

https://github.com/aalsanie/codes
4•aalsanie•6d ago•0 comments

Roget's Thesaurus

https://artflsrv04.uchicago.edu/roget-thesaurus/
19•gregsadetsky•1d ago•2 comments

Smartphone LED detects hidden cameras with AI

https://www.chosun.com/english/industry-en/2026/08/30/SBFXUIJQYZEARKP5T4FBAY25HQ/
223•geox•2d ago•67 comments

Flat vs. segmented memory – it's recursive

https://www.humprog.org/~stephen/blog/2026/08/25/#flat-vs-segmented
27•signa11•5d ago•0 comments

Lion-man

https://en.wikipedia.org/wiki/Lion-man
120•gurjeet•2d ago•65 comments

DoltLite: A SQLite fork with Git-style version control, built with 2k agent PRs

https://www.dolthub.com/blog/2026-08-31-doltlite-beta/
52•lbw1215•10h ago•36 comments

Reverse engineering my ADHD test

https://nullpt.rs/reverse-engineering-adhd-test
234•hazebooth•4d ago•142 comments

RavynOS: Pre-alpha open-source OS based on Darwin, FreeBSD, Apple open-source

https://ravynos.com/
205•Bluestein•19h ago•115 comments

Borges Labyrinth in Venice reopens to the public

https://www.wallpaper.com/design-interiors/labirinto-borges-venice-reopening
79•gone35•3d ago•28 comments
Open in hackernews

44% on ARC-AGI-1 in 67 cents

https://mvakde.github.io/blog/44-on-arc-1/
65•porridgeraisin•1h ago

Comments

embedding-shape•1h ago
Is the author only running their model against one benchmark? I don't think anyone finds that difficult to achieve, the difficulty comes when you want to make the model not benchmaxxed to a specific benchmark, and generalize so it can solve problems not part of the training data, but seems this model is specifically for not this? How useful is that?

If you just wanted to pass these specific tasks in this specific benchmark, and wanted to do so cheaply, I'm sure a non-LLM-based approach would yield better results for even cheaper, since what the author's model does, seem to basically be "solve ARC puzzles", not a general LLM or "coding" LLM.

f311a•42m ago
The whole point of his model is to optimize for a very specific benchmark.

BUT, he does not use labels when training, so the model does not know the answers.

embedding-shape•33m ago
> The whole point of his model is to optimize for a very specific benchmark.

But benchmaxxing is what we generally try to avoid for training, as there is no point really for it. We used to call it "overfitting", now you're saying this person does it intentionally? Why?

f311a•27m ago
Why not? There is $700k reward for the next iteration of this benchmark https://www.kaggle.com/competitions/arc-prize-2026-arc-agi-2...

I would not call this overfitting, it's finetuning for specific task where you have a benchmark.

K0balt•10m ago
There are plenty of applications where a machine learning system needs to optimize for a very limited data set that is still intractable by linear logic systems of reasonable scale and complexity. It’s interesting, because he is using the legos of LLMs to build highly specialized machine learning systems, which is a very pragmatic approach. Obviously a lot of other ways to achieve similar goals, but it’s cool to see someone back porting the modern tools towards older style optimizations.

Also, the complexity of the task he is using occupies an interesting middle ground of ultra high dimensionality (for a “simple” problem) while being limited in width to a narrow set of solves- a space where one would be tempted to imagine you would need a much more capable system.

bkaae•39m ago
I read this as a response to the current hype around LLMs. He is showing computers can solve these issues, without using an LLM architecture. A lot of people have sort of forgot that machine learning is more than just LLMs these days.

I found it to be a very interesting angle.

embedding-shape•34m ago
> He is showing computers can solve these issues, without using an LLM architecture.

Isn't it a LLM he's building though? My very point is that this particular use case could be solved better without building a LLM, now you claim he is not? The description of what he's doing surely makes it sound like it's a (very small) LLM, and personally I'm still on the "if it quacks like a duck" train in life.

> A lot of people have sort of forgot that machine learning is more than just LLMs these days.

Yeah, which I guess if you make my previous comment more concise, is exactly what I state too.

dakolli•8m ago
Nowhere does he say he built an llm. Hes using a transformer, not an llm.
embedding-shape•4m ago
> Nowhere does he say he built an llm. Hes using a transformer, not an llm.

Please describe what in your mind a "LLM" is exactly, then describe what this person is building. To me this sounds like "He's not building a calculator, he's just building a program that can do addition, minus, multiplication and division and display the results".

Obviously it's not a Large Language Model, but to me this looks more like a LLM than not, given the architecture he's chosen. But again, maybe I misunderstand?

xeonax•1h ago
Even cooler is his about me mention of saving his own life https://mvakde.github.io/ > Saved myself in a medical emergency (doctors didn't know what rhabdomyolysis was)
qlm•25m ago
Crazy, considering rhabdo isn't that rare.
p-e-w•18m ago
“What do you call a medical student who graduated at the bottom of their class?”

“Doctor.”

eis•49m ago
> Increases in LLM scores are now mainly driven by post training (evidence in next section) and are probably a function of amount of synthetic data. They are learning to solve ARC tasks, not learn general abstract reasoning

Agreed and that's for any benchmark. Private tests are better but you still have to trust the provider to not log and use them for training.

That's why I like when a new set of tests like a new ARC-AGI version is published, that's where you can see which of the models abstracted to more general capabilities instead of being focused on the previous tasks. Most models completely fail new ARC-AGI tests.

The "67 cents" part though is misleading imho. You can't extrapolate from there and think that investing say $100 will get you a lot better results. You hit a ceiling very fast and investing into more compute will give you diminishing results. So yes, you can train a custom model to do somewhat decently on a specific set of tasks but then what?

bkaae•37m ago
Then nothing - that's awesome. People think that LLMs are the know-all do-all solution to every problem now.

Putting solutions in terms of cents is a great way to potentially win over some ai boosters imo. There are other ways to solve hard problems.

larodi•22m ago
"I don’t understand why others didn’t figure this out"

- how about we allot the possibility that so many of presumed ML experts don't have any clue what they be doing, and are eventually API bitches, nothing more.

pwmglenn•18m ago
Really impressive and creative research. I wonder if the leading labs do anything similar with their models? It doenst look like the open source labs do?
kvn95ss•5m ago
> Also, I’m not sure whether “general reasoning” even exists in the first place? Maybe humans are specialised too

I have been wondering the same. We are now exposed to so many stimuli, we are tricked into thinking this is the norm - to have a reasonable understanding about everything, unless specialization is called for.

evilmathkid•5m ago
Hi! Author here. Surprised to see this on HN now. Happy to answer any questions!

Some context about this:

- This is NOT an LLM. its a small ar transformer trained from scratch. One of the points was that extremely complex problems can be tackled without LLMs

- Till the v1 of this result, this benchmark was only scaled by LLMs or their finetunes (ofc w enormous training costs). Other attempts performed okayish but used v complex architectures or extremely high amounts of training compute. No one expected a simple AR transformer to perform this well, at this low cost and w these few training samples.

- Sample Efficiency is one of the most important unsolved problems today in AI. That's what I was targetting with this work. We know it is easy to increase SE by increasing compute/params, so it was important to constrain cost as much as possible (also why OpenAI's Parameter Golf had fixed compute and why Modded NanoGPT is considered very sample efficient)

- Can the perf be improved? Yes but the competition is ongoing so can't talk about it

- Personally I think today's frontier models can be beat by training from scratch. Haven't proved this yet tho

- Fun: I was new to ML when I posted this first (dec '25). I basically used ARC as a way to learn ML