snats

11 ms·

31.
Tokin: Internet's Best Videos(tokin.tv)
2 points by snats 2y ago|1 comments
32.
by snats 2y ago
just changed! thanks for the heads up
33.
I want to break some laws too(snats.xyz)
14 points by snats 2y ago|2 comments
34.
In place of a legal notice, here is a blessing(sqlite.org)
3 points by snats 2y ago|0 comments
35.
by snats 2y ago
hi! i used the average accuracy over the entire dataset made originally made by the llm
36.
by snats 2y ago
Hi! Author here, I wasn't expecting this to be at the top of HN, AMA
37.
by snats 2y ago
Is there any goodmodel for OCR but on handwritten information? I feel like most models are currently kind of trash
38.
Infinifi Cafe: infinite lo-fi music in the background(infinifi.cafe)
2 points by snats 2y ago|0 comments
39.
by snats 2y ago
Deeptech is going crazy, all of the AI boom is pretty much an obvious subject if you are in media, and finally you still have your usual gossip. You still have a lot of eyes
40.
by snats 2y ago
You can do the second part that is C!
41.
Shallow Theory(snats.xyz)
3 points by snats 2y ago|0 comments
42.
Differential Evolution and Clip = Image Generation(snats.xyz)
1 points by snats 2y ago|0 comments
43.
New attention mechanisms that outperform standard multi-head attention(arxiv.org)
233 points by snats 2y ago|49 comments
44.
You Can Optimize Anything(snats.xyz)
1 points by snats 2y ago|0 comments
45.
Sakuga-42M Dataset: Scaling Up Cartoon Research(arxiv.org)
85 points by snats 2y ago|31 comments
46.
PyTorch Neat(github.com)
1 points by snats 2y ago|0 comments
47.
by snats 2y ago
Remember that Llama400B is still in training and could end up crushing a lot of the use cases for GPT-4
48.
by snats 2y ago
I recently wrote a writeup on bigrams and the infinigram outputs[1]. I genuinely believe that ngrams are making a comeback. For query searching it's really good. [1] https://snats.xyz/pages/articles/from_bigra
49.
From Bigram to Infinigram(snats.xyz)
1 points by snats 2y ago|0 comments
50.
From Bigram to Infinigram(snats.xyz)
1 points by snats 2y ago|0 comments
51.
Curse of Dimensionality(en.wikipedia.org)
11 points by snats 2y ago|0 comments
52.
by snats 3y ago
A bit off topic but every time I see DTrOCR I remember that marketing is a good idea because DTrOCR[1] and Fuyu[2] are basically the same architecture[3]. [1] https://arxiv.org/pdf/2308.15996v1.pdf [2] https:/&#x
53.
by snats 3y ago
It's been pretty interesting to see some research trying to solve this. For example, Stanford researchers recently published QuietSTaR[1]. This makes LLMs "think" before they speak. By generating a chain of thoughts and just
54.
by snats 3y ago
to this day i am impressed that they have not figured out how to embed advertisments to bard outputs so that they can go free.
55.
by snats 3y ago
No! I only tried PCA, but I still have the embeddings. I'll try later and post results.
56.
by snats 3y ago
I built a map of all the PDF urls on the internet recently. I used a tiny embeddings model and PCA for dimensionality reduction. https://weblog.snats.xyz/posts/2024/03/20/
57.
Tell HN: Ai.com redirects to Google Gemini now
4 points by snats 3y ago|2 comments
58.
Tiny Model PDF Corpus: URL Embeddings for 8.3M PDFs(weblog.snats.xyz)
1 points by snats 3y ago|0 comments
59.
Same Architecture, Nicer Presentation(weblog.snats.xyz)
1 points by snats 3y ago|0 comments
60.
UTF-8 Everywhere(utf8everywhere.org)
2 points by snats 3y ago|0 comments

More ›