heyitsguay

8 ms·

1.
by heyitsguay 16d ago
Creating a basic ergonomic example of the full train/deploy loop, using your data engine, for a popular platform in the zeitgeist, will give you a traction boost. You don't have to innovate on the parts outside your core competenc
2.
by heyitsguay 1mo ago
Yeah similar to how image generators don't draw Mickey Mouse, but ask for a black cartoon mouse with big ears and see what you get!
3.
by heyitsguay 2mo ago
Products are something other than the sum of their technical parts.
4.
by heyitsguay 3mo ago
> This gestalt isn't just for biological organisms, but any system for which its decision making engages with representations of the external environment unified with a self-representation to form a coherent representation of a pers
5.
by heyitsguay 3mo ago
The two outputs (an image and a string sequence) are learned from entirely different data, even when they're generated with the same model, which is not always the case. There's nothing intrinsic to the learning process that conne
6.
by heyitsguay 3mo ago
It was the same sniper that blew up that Starship
7.
by heyitsguay 6mo ago
What were the results of 3d fractal shader pretraining?
8.
by heyitsguay 7mo ago
Interesting, can you explain why?
9.
by heyitsguay 7mo ago
Is it inflammatory if it's true? Seems rather embarrassing (but fitting for 2026 America) to shy away from painful truths because they're uncomfortable.
10.
by heyitsguay 8mo ago
Moloch!
11.
by heyitsguay 8mo ago
Thanks for this answer! I appreciate the clarity, I can see the economic impact for your company. Very cool.
12.
by heyitsguay 8mo ago
Agreed. I've never seen a concrete answer with an outcome that can be explained in clear, simple terms.
13.
by heyitsguay 11mo ago
I went through a similar cycle. Going back to simplicity wasn't about laziness for me, it was because i started working across a bunch more systems and didn't want to do my whole custom setup on all of them, especially ephemeral s
14.
by heyitsguay 1y ago
What problems?
15.
by heyitsguay 1y ago
Pass a law requiring cloud compute providers to accept a maximum user budget and be unable to charge more than that, and see how quickly the big cloud providers figure it out.
16.
by heyitsguay 1y ago
What do you do with agents?
17.
by heyitsguay 1y ago
Are there any? Concretely. Genuinely curious.
18.
by heyitsguay 1y ago
"Unfortunately, my content guidelines prohibit me from describing my activities with your mother last night"
19.
by heyitsguay 1y ago
It's the same amount of effort benchmarking, just a better choice of backbone that enables better choices of benchmark tasks. If the claim is that a Tversky projection layer beats a linear projection layer today, then one can test whet
20.
by heyitsguay 1y ago
Seems cool, but the image classification model benchmark choice is kinda weak given all the fun tools we have now. I wonder how Tversky probes do on top of DINOv3 for building a classifier for some task.
21.
by heyitsguay 1y ago
Do you have references on adaptive methods for image recognition?
22.
by heyitsguay 1y ago
I've done something along these lines! https://github.com/heyitsguay/trader The challenge for me was consistency in translating free text from dialogs into classic, deterministic game state changes. But what'
23.
by heyitsguay 1y ago
Object detectors output detection bounding boxes along with confidence scores. The higher the score, the more confident the model is that the associated bounding box is a correct detection. When used in applications (like this one), the use
24.
by heyitsguay 1y ago
I've noticed this a lot, too, in HN LLM discourse. (Context: Working in applied AI R&D for 10 years, daily user of Claude for boilerplate coding stuff and as an HTML coding assistant) Lots of "with some tweaks i got it to work
25.
Strategic Overclaiming of LLM Reasoning Capabilities Through Evaluation Design(huggingface.co)
1 points by heyitsguay 1y ago|0 comments
26.
by heyitsguay 1y ago
Defense needs them everywhere, offense needs one gap in the defense.
27.
by heyitsguay 1y ago
At Allen Control Systems, we're working on deploying automated anti drone turrets (radar or passive EO detection) right now! Development in the space is happening at a breakneck pace. We're hiring pretty aggressively, if this sort
28.
by heyitsguay 1y ago
Fwiw, doctor - woman + man = medical practitioner too
29.
by heyitsguay 1y ago
There are many, many papers and projects out there about tuning foundation models on various types of medical imaging data, and many organized efforts to produce large medical imaging datasets to feed that training. This stuff is well-known
30.
by heyitsguay 1y ago
I ran a comparison of DINOv2 with and without registers on some image embedding tasks for work; DINOv2+registers saw a performance metric bump of 2-3%. Not nothing, not transformative, worth using when the only difference for inference is t

More ›