xucheng

8 ms·

1.
NeurIPS bans submissions from US-sanctioned institutions(neurips.cc)
3 points by xucheng 6mo ago|1 comments
2.
by xucheng 6mo ago
> Sanctioned Institutions > The NeurIPS Foundation, like any entity operating within U.S. legal jurisdiction, is required by law to comply with U.S. sanctions and trade restrictions. Under these regulations, providing 'services&#
3.
by xucheng 1y ago
It’s quite common for a DOI to be assigned to a paper after it’s accepted during camera ready. However, the DOI won’t work until the conference or journal version is published on the official website (ACM in this case). The version you’re v
4.
17 Pages [video](nebula.tv)
2 points by xucheng 1y ago|0 comments
5.
by xucheng 1y ago
Can this be solved by storing a timestamp of the record along with precise GPS coordinates? Could we then utilize some database to compute the drift from then and now?
6.
Nebula Sans(nebulasans.com)
348 points by xucheng 1y ago|101 comments
7.
by xucheng 2y ago
At least, Cisco survived. Nortel and Lucent were not so lucky.
8.
by xucheng 2y ago
Many years ago, I reported an issue where iTerm2 leaks sensitive search history to preference files [1]. The issue was quickly fixed. But until this day, I can still find people unintentionally leak their search history in public dotfiles r
9.
by xucheng 2y ago
Interestingly, in the end of the article, the FDA links to an old article hosted on web.archive.org[1] even though the linked article was originally published by FDA themselves. Considering the linked article was only published at 2022, a m
10.
by xucheng 2y ago
Maybe better url, the official press release : https://www.gilead.com/news-and-press/press-room/press-relea... @dang
11.
Framework for Legislating on Artificial Intelligence (2023) [pdf](eff.org)
2 points by xucheng 2y ago|0 comments
12.
by xucheng 3y ago
A related question: what is the state in term of supporting the SQL standard among the popular RDBMS? It seems that almost all database engines use their own custom syntax.
13.
by xucheng 4y ago
You are confusing two different units. The earthquake moment magnitude scale measures the amount of energy of a given earthquake. It does not concern the location you are and the shacking you felt. Therefore, for this particular Taiwan eart
14.
by xucheng 4y ago
I mean that it is hard to scrape these git.io links used in the research papers to build the archive. Unless of course, if Github provides a DB dump, it would help everyone a lot.
15.
by xucheng 4y ago
It is unlikely because it is very difficult and maybe illegal to scrape research papers. See https://en.wikipedia.org/wiki/Aaron_Swartz for example.
16.
by xucheng 4y ago
List of takedown notices from HackerRank to GitHub: https://github.com/github/dmca/search?q=HackerRank&type=
17.
by xucheng 4y ago
> figure out a way where we can do things like randomizing questions but preserving the integrity of it to ensure a fair evaluation, etc How does any of these have anything to do with copyright infringement in the context of DMCA takedow
18.
by xucheng 5y ago
> The wiki listing of packages acts as the "registry" and populates the clib-search(1) results. This seems to be an extremely bad idea. Since everyone can edit the GitHub wiki pages, it means its registry is vulnerable to all k
19.
by xucheng 5y ago
> Importantly, you must do this for every open branch in your repo. It is not enough to do so for only your default branch since a malicious PR can target any of your open branches. That is, if you have an open branch that uses a vulnera
20.
by xucheng 5y ago
Compared with Rust’s Result monad, which allows developers to clearly see the effects of error handling, there are two other hidden fallible effects in Rust that are much harder to tackle: * Panic rewinding. I am not sure how to ensure your
21.
by xucheng 5y ago
May I ask what missing fields you are referring to? Why @online/@software/@dataset type in biblatex [1] cannot do the job? That being said, I think GitHub should acknowledge that it is common for authors to want people cite their
22.
by xucheng 5y ago
Since many papers have existing bibtex, it would be better if GitHub either provides tools to convert bibtex to CFF or supports bibtex directly. FWIW, bibtex is the de facto standard.
23.
by xucheng 5y ago
In addition to the attacks, such as converting legit image to be detected as CSAM (false positive) or circumventing detection of the real CSAM image (false negative), which have been widely discussed in HN, I think this can also be used to
24.
by xucheng 5y ago
Just like any autoencoder, it is not about getting back the exact original, which is of course impossible. It is about summarizing the image in 96bits information, which is quite enough to leak the gist of the original image. For example, [
25.
by xucheng 5y ago
> 4. Actually, need to run step 1-3 at least 30 times Depending on how the secret sharing is used in Apple PSI, it may be possible that duplicating the same image 30 times would be enough.
26.
by xucheng 5y ago
You can technically hide an adversarial collision inside a complete legit normal image. It won’t be seen by human eyes but it will trigger a detection. In addition, you can do the complete opposite by perturbing a CSAM to output complete di
27.
by xucheng 5y ago
In addition to generate the adversarial collisions, someone mentioned that it can also be used to train a decoder network to reverse any NeuralHash back to its input image.
28.
by xucheng 5y ago
> Neural hash generated here might be a few bits off from one generated on an iOS device. This is expected since different iOS devices generate slightly different hashes anyway. The reason is that neural networks are based on floating-po
29.
by xucheng 5y ago
See https://news.ycombinator.com/item?id=28105849 , which shows a POC to generate adversarial collisions for any neural network based perceptual hash scheme. The reason it works is because "(the network) is continuous(l
30.
by xucheng 5y ago
Reddit discussion: https://old.reddit.com/r/MachineLearning/comments/p6hsoh/p_a...

More ›