ya3r

31 ms·

1.
by ya3r 5mo ago
Telegram secure chat messages do this by default.
2.
by ya3r 6mo ago
Are we aware of any attacks (or claims of attacks) against any previous version of the iPhone's Lockdown mode?
3.
by ya3r 1y ago
The regular copilot, copilot.microsoft.com
4.
by ya3r 1y ago
Have you seen Microsoft's copilot? It is essentially free openai models
5.
by ya3r 3y ago
Maybe relevant blogpost is https://devblogs.microsoft.com/python/python-311-faster-cpyt... dated Oct 2022. The team behind this and some other recent improvements to Python are at Microsoft.
6.
by ya3r 3y ago
Bicycle by Bartosz Ciechanowski: https://ciechanow.ski/bicycle/ - HN post: https://news.ycombinator.com/item?id=35343495 Bartosz has many great blog posts, maybe most famously the one on mechanical wat
7.
by ya3r 3y ago
> "As a reminder, there is not a single LLM on the market today that is not vulnerable to prompt injection ... And by and large, companies don't really seem to care." So far from the truth. I know that there are entire tea
8.
by ya3r 4y ago
It only works, if it consistently works. And it only doesn't work, if you tried it consistently and it didn't work.
9.
by ya3r 5y ago
tqdm is really great. (Also great name BTW) It has a lot of features like cli, jupter notebook widget, flexibility, etc. I have been using it for 4-5 years now.
10.
by ya3r 5y ago
> Based on our research and findings, we assess this to be one of the most technically sophisticated exploits we've ever seen, further demonstrating that the capabilities NSO provides rival those previously thought to be accessible
11.
by ya3r 5y ago
Definitely a thing. Regarding the cost, it might be more than "some paperwork".
12.
by ya3r 5y ago
Nat was a great CEO. The best that could have happened to Github after the acquisition IMO. The one time that I won't forget about Github under Nat, was when they stood up for Iranian developers [1]. They went the extra distance to get
13.
by ya3r 5y ago
After the new name fails to make any significant difference, they will try new CEO.
14.
by ya3r 5y ago
Technically speaking, this does not prove that an adversarial attack is possible on the CSAM system of apple, Given that apple has another not released neural hash system on their servers which is potentially larger and works better than
15.
by ya3r 7y ago
I really like the fastai abstractions and their attention to detail. Also their callbacks are amazing. I always look at their implementations for inspiration. But my main issue with it is the code formatting ( https://docs.fast.ai
16.
by ya3r 7y ago
This reminds me of this great video: https://www.youtube.com/watch?v=iHzzSao6ypE
17.
by ya3r 8y ago
First off, as many know, this is not a single incident. This has happened to many of us (including me) from "sanctioned" countries. (I am from Iran, but live in Germany) But this makes me think. Many of the services I use are from
18.
GitHub Discussion – Optimizer Unification in TensorFlow 2.0(github.com)
2 points by ya3r 8y ago|0 comments
19.
SSPipe: Simple Smart Pipe for Python(github.com)
7 points by ya3r 8y ago|0 comments
20.
by ya3r 8y ago
Convex Optimization by Stephen Boyd. https://lagunita.stanford.edu/courses/Engineering/CVX101/Win...
21.
by ya3r 9y ago
I have used similar setups in Sublime Text for Python development, the main issue is instability and inefficiency. Installing these set of packages in sublime makes things slower and more unstable. Multiple times while editing large files m
22.
KaTeX: Fast math typesetting for the web(github.com)
1 points by ya3r 10y ago|0 comments
23.
by ya3r 10y ago
One thing which is missing is a "model zoo", a place where people share models (usually well-known) pre-trained, which is very useful for starting to use a framework. Although this repo https://github.com/baidu&#x2
24.
by ya3r 10y ago
In my field (Computer Vision/Machine Learning) newest research papers usually get into arXiv before getting accepted in any conferences. So I try to keep up with the arxiv's rss of this field. Further more I follow other people in
25.
by ya3r 10y ago
You are right. I watched them on youtube.
26.
by ya3r 10y ago
The videos have very poor quality. The talks don't contain recording of the slides. IMHO the only one worth watching and actually is interesting is the discussion panel.
27.
by ya3r 10y ago
I guess this (switching from Torch to other deep learning libraries) will become a trend as deep learning have become more mainstream in tech companies. I say Facebook, Twitter and others who use Torch (I don't know of any others actua
28.
Paymentwall opens up for Iran(blog.paymentwall.com)
1 points by ya3r 10y ago|0 comments
29.
by ya3r 11y ago
This method achieves a new state of the art performance on ILSVRC2012 top-5 with 3.46%.
30.
by ya3r 11y ago
To me this shows how similar Tensorflow and Theano are in their APIs.

More ›