cowl

17 ms·

151.
by cowl 3y ago
>If I were to start with ChatGPT I would, necessarily, only say what has been said. You would but not chatGPT. I have been very vocal about cautioning people that LLM are just a parrot, but that is in the sense that there is no real mea
152.
by cowl 3y ago
easier done with dedicated Controllers instead of one BIG controller that needs to containerise its software? Why does the rear camera and lights need to use the same controller as the Engine sensors? This way you even avoid the latest &quo
153.
by cowl 4y ago
No it's not, you cleary are misinterpretting the article. >we defined an emergent ability as an ability that is “not present in small models but is present in large models.” models are taken (because there is no other way) as black
154.
by cowl 4y ago
Where do you see this kind of Meaning in physics? In physics we have of example the Emergence of fluid dynamics and is clearly defined as a behaviour or property of system that does not derive simply from the sum of combination of it'
155.
by cowl 4y ago
not neccessary a new discovery just a new behaviour for which it was not trained on. See https://en.wikipedia.org/wiki/Emergence a classic example is structure of a flock of starlings in flight or a school of fish, t
156.
by cowl 4y ago
> in this work, we identify and study four key stages for successfully teaching algorithmic reasoning to LLMs: (1) formulating algorithms as skills, (2) teaching multiple skills simultaneously (skill accumulation), (3) teaching how to co
157.
by cowl 4y ago
Maybe but since we can't by definition know what is present in the model we can not define any behavior as emergent as opposed to simply trained. Suppose you don't know anything about our school system and you observe that 12-gra
158.
by cowl 4y ago
in fact their example of https://github.com/google/BIG-bench/tree/main/bigbench/bench... showed 0 emergent behavior in understating modified arithmetic even in large LLMs. the accuracy (and lack of
159.
by cowl 4y ago
an Emergent behavior is usually defined as a behavior that is not expected from the training or the starting configuration. Their Emergent Definition is absurd. is like saying that understanding Calculus is emergent behavior in 12 grade be
160.
by cowl 4y ago
The Effect of that kind of policy was already tried in Italy to a certain extent. The result was that For every new "Temporary Investment" (meaning a project that needed 1-3 years of pushing) Noone was hiring and everyone, no exc
161.
by cowl 4y ago
Or companies will simply use the trial period to terminate. Painful as it is it's much better to get a revocation of offer than for the company to go through the motions just to arrive at the "clear" moment of trial period a
162.
by cowl 4y ago
What you missing is that by chosing this obvious profit center they risk a much larger profit center because the backlash. It's not a moral thing, it's a calculated choice. That's why who takes this risks also charges a much
163.
by cowl 4y ago
Don't look fundamentally different? May I'm missing something but I have not seen a single tutorial or article about something Debian or Firefox that starts that with the disclaimer "This is not affiliated with the official D
164.
by cowl 4y ago
Unfortunetly less and less apps nowdays can work in an exclusive offline mode. There is no way to distinguish between legitimate traffic and "telemetry" trafic.
165.
by cowl 4y ago
Well Everyone is reporting it. And OpenAI say that they use the prompt to improve the model: > When you use our non-API consumer services ChatGPT or DALL-E, we may use the data you provide us to improve our models. You can request to op
166.
ChatGPT leaking Samsung chip secrets is iceberg’s tip(eenewseurope.com)
10 points by cowl 4y ago|3 comments
167.
by cowl 4y ago
The real problem lies in the fact that those non existing citations will become real. Several years back there was a case of an article in Wikipedia that made some unsrourced claims->Journalist that doesn't do verification republis
168.
by cowl 4y ago
The problem is not data-collection per se. optout data collection is the problem. if the user is informed and consents there is no problem at all.
169.
by cowl 4y ago
In fact the article is wrong. Doesn't matter what data they do collect, they need to establish a connection to the telemetry server and that exposes the IP (which is Personally Identifiable information) without the user's consent
170.
by cowl 4y ago
Optout telemetry is illegal in Europe. Any connection that is not neccesary for the purpose of the tool is exposing the IP of the user without his consent. Pinky promises that IP of the connection will not be (ab)used do not count. See htt
171.
by cowl 4y ago
> Assuming no personal identifyable information is saved, this is legal. This is not true in Europe Anymore. Any Optout telemetry is illegal because they have to establish a connection to the telemetry server and thus leaking the user I
172.
by cowl 4y ago
> Because fingerprinting. I can't think of any way to stop fingerprinting. The best I can do (and I do) is to disallow JS from running, but that still leaves many possible signals. This is even worse for native apps. webapps do fing
173.
by cowl 4y ago
> not enough context, maybe > Probably some corporate compliance drones and underfunded government stooges. Youtube alone has 720,000 hours of fresh video content per day Tiktok must have less number of hours but more Videos per day.
174.
by cowl 4y ago
The problem is what is an Ad? Is the Cryptobro shilling for the coin of the week an Ad? is the Influencer that "casually" is using her brand of Mascara while talking an Ad? Do creators need to have two versions of the same video (
175.
by cowl 4y ago
Do you have any example in mind? The three "main competitors" right now are AnyCubic, Creality, and Bambu. All with their own Designs (and various levels of contributing back) that have nothing in common with Prusa appart the fact
176.
by cowl 4y ago
Does this remote usecase justify it being a bultin feature of every car instead of a dedicated device that only the people that need it would buy?
177.
by cowl 4y ago
Yes I said there are patents. Your protection does not come from the license in this case. The license in this case it's just a contractual feature. Your protection comes from the patent. IE, if someone uses your patent technology you
178.
by cowl 4y ago
Yes you are right that are a kind of threat response but not personal threat as much as general threat. There is a great danger that because of the name AI and the general association that we have with computers as giving correct result, Pe
179.
by cowl 4y ago
The argument of Stochastic Parrot is not that we shouldn't do it. It's just "Do not attribute any meaning to it". The one in this article is in the same vane that author thought the AI was learning to program when in re
180.
by cowl 4y ago
Well what you are saying is that the argument of the creators of the model is erroneous and you know better. More power to you.

More ›