JacobiX

10 ms·

61.
by JacobiX 4y ago
My understanding about the date of training from ChatGPT page: ChatGPT is fine-tuned from a model in the GPT-3.5 series, which finished training in early 2022. So it was based on a model from early 2022 but it was partially trained (fine-tu
62.
by JacobiX 4y ago
Not sure if they are new, I searched the first one and the first Google search result contains a solution for it: https://puzzlehunt.club.cc.cmu.edu/protected/solutions/21033...
63.
by JacobiX 4y ago
The problem with many of the tasks that people are trying is: the answers are already available on the internet for those very popular crosswords. For example a quick search for "1 Chap recalled skill: something frequently repeated (6)
64.
by JacobiX 4y ago
It wasn't clear from the title that this is a db designed for embedded devices (unlike for example SQLite that belongs to the family of in-app embedded databases). One of the interesting features in this db is wear leveling a technique
65.
by JacobiX 4y ago
I love those tiny DNN frameworks, some examples that I studied in the past (I still use PyTorch for work related projects) : thinc.by the creators of spaCy https://github.com/explosion/thinc nnabla by Sony https:&#x2F
66.
by JacobiX 4y ago
Not sure about the cheaper model part, because one cannot transform trivially a big SOTA langage modeling models into a digital assistant. So maybe they are not using a cheap ML model after all ... Also, for a digital assistant the input is
67.
by JacobiX 4y ago
The fact that a machine learning model can solve a seemingly hard task A does not automatically translate to solving a lesser hard task B. Conversational AI that understand the intent remains a really hard problem even if speech recognition
68.
by JacobiX 4y ago
Thanks, I've replied privately
69.
by JacobiX 4y ago
We used it in a project that performed document recognition and analysis. It was useful for aligning the fields to be extracted from the source image as we formulated it as an optimization problem. It was straightforward to use and it conta
70.
by JacobiX 4y ago
The author talks about AI management solutions that will reduce management head count as if such solutions are available and ready to use. But I'm not aware that such solutions are available yet.
71.
by JacobiX 4y ago
I'm not so sure about that. Of course computers can do arithmetic operations, but this is not the same as solving math problems, proving theorems, etc. Even mathematical objects are approximated up to an approximation error in a compu
72.
by JacobiX 4y ago
Hope that they will extend this regulation to not just mobile phones, tablets and cameras but also to electric razors, hair care appliances, toothbrushes, home appliances, etc ... considering that many big EU corporations manufacture those
73.
by JacobiX 4y ago
It would be nice if those same EU regulations where applied to all industries and not some specifically designated ones. For example my German Braun Shaver Charger will not be impacted by this regulation ...
74.
by JacobiX 4y ago
I’ll just use xxHash family, murmur hash family or even FNV. For crypto, BLAKE2 is a cryptographic hash that’s faster than SHA,MD5, etc and is at least as secure as the SHA-3.
75.
by JacobiX 4y ago
Yes, because SHA256RNDS2 aid the implementation of the update function of SHA256. We should invoke it multiple times on the message blocks, it performs two rounds out of the 64 SHA256 rounds.
76.
by JacobiX 4y ago
Not the OP, but those extensions implements cryptographic hash functions, they are not very suitable (too slow) for hash-tables and other hash based data structures.
77.
by JacobiX 4y ago
It would be great if it supports the LC3 codec (Bluetooth’s future low-power and high-quality codec)
78.
by JacobiX 4y ago
At first, you can experiment with "extreme" input that can cause overwriting like: a photo of Marylin Monroe, etc. To search using reverse image search I use yandex, and I downsample the image.
79.
by JacobiX 4y ago
The first thing that I try after generating an image from DALL-E is using reverse image search. I do it on every image that I intend to use, more often than not, I find a very similar image, in this case I discard it and vary my prompts.
80.
by JacobiX 4y ago
> The headline is "Gas crisis sends euro back below parity against dollar." > Yet the article does not explain how a "gas crisis" leads to a weaker currency. It wasn’t clear (at least for me) why gas has such an im
81.
by JacobiX 4y ago
We use Azure and Linode for all of our cloud services. The number one reason: we can contact human support at any time and we have a designated customer success manager. In many situation this proved to be critical, especially for resolving
82.
by JacobiX 4y ago
A better title would have been: Linux performance under Apple M2 vs AMD Ryzen 7 Pro 6850U.
83.
by JacobiX 4y ago
This is the first time I hear about memgraph, its really interesting because I'm beginning a project that involves the analysis of some large graphs. Sometimes, before using a lib, I dig into the GitHub repo. So here, I viewed their i
84.
by JacobiX 4y ago
I'd love to get a look at the source code. Alas, the code is unavailable!
85.
by JacobiX 4y ago
As I understand from reading Vaclav Smil's “How the World Really Works”. Replacing fossil fuels by renewables isn't simple: "we are a fossil-fueled civilization whose technical and scientific advances, quality of life, and pr
86.
by JacobiX 4y ago
> Does this suggest we can expect enforcement of IP protection to fade everywhere someday soon? It could also be the other way around: in order to expand to new markets, Chinese companies will be obliged to comply with patents/copyr
87.
by JacobiX 4y ago
One outstanding fact is the disproportionate number of fields medalist that went to the École normale supérieure in Paris !
88.
by JacobiX 4y ago
Interesting when testing with inputs like "Oscar Wilde photo" or "marilyn monroe photo" and comparing to a Google image search. After some iterations we can have quite similar images but the faces are always blurry.
89.
by JacobiX 4y ago
It’s the same problem with those ML models, the other day someone generated a children’s book using GPT3, turned out that there is a real children's book with the same name and a very similar content: The Very Lonely Firefly by Eric C
90.
by JacobiX 4y ago
For me the book that changed my perspective on proofs was : Proofs from THE BOOK by Martin Aigner and Günter M. Ziegler. I always thought that proofs are just means to an end, but this book showed me that proofs can be more interesting than

More ›