j2kun

10 ms·

151.
by j2kun 9mo ago
The author calls the mistakes "confirmed hallucinations" without proof (just more or less evidence). The data never "speak for itself." The author curates the data and crafts a story about it. This story presented here i
152.
by j2kun 9mo ago
I spot-checked one of the flagged papers (from Google, co-authored by a colleague of mine) The paper was https://openreview.net/forum?id=0ZnXGzLcOg and the problem flagged was "Two authors are omitted and one (Kyle Ric
153.
by j2kun 9mo ago
I saw that huge box of decks they printed for this, and I though, oh dear, how are they going to sell that many copies? :(
154.
by j2kun 9mo ago
Make a social app whose goal is to get people off their phone as quickly as possible. There used to be a slew of apps where you a press a button to indicate "I'm bored/free, who wants to hang out?" and then you get match
155.
by j2kun 9mo ago
A theory besides... not enough training data? Is it even possible to formulate a coherent theory about this? I'm talking about customizing a widely-used build system, not exactly state-of-the-art cryptography. What could I possibly say
156.
by j2kun 9mo ago
I think you should read the article again, because this comment is a straw man vis-a-vis the article.
157.
by j2kun 9mo ago
Sounds like you should go bundle sub-prime mortgages into some complex securities, if you like intentionally living on the knife's edge of disaster.
158.
by j2kun 9mo ago
How would one set this sort of test up? I surely have example domains where LLMs routinely do poorly (for example, custom bazel rules and workspaces), but what would constitute a "showcase" here?
159.
by j2kun 9mo ago
The author claims it's not just that one evangelizes it, but that they become hostile when someone claims to not have the same experience in response. I don't recall Either Willison or Antirez scaring people by saying they will be
160.
by j2kun 9mo ago
I do wish I could have a good, in-depth tutorial on how to set this up myself. Along with (pipe dream) an explanation of how it would interact with my local utility. I worry that due to some silly technicality, I won't be able to expor
161.
by j2kun 9mo ago
They are good for a jump start on literature search, for sure.
162.
by j2kun 9mo ago
> A large amount of Tao's work Perhaps you meant, "a decent amount of his recent work." He has been doing math long before LLMs, and is still regularly publishing papers with collaborators that have nothing to do with AI.
163.
by j2kun 9mo ago
Would be useful to have the website say something, _anything_ about what this is doing besides asking you to upload an image.
164.
by j2kun 9mo ago
This is the kind of choice an LLM would make...
165.
by j2kun 9mo ago
Doesn't it have sensors too? I recall reading about proximity sensors to other smart bricks, as well as accelerometer and orientation sensors.
166.
by j2kun 9mo ago
https://michaelhalassa.substack.com/p/llm-induced-psychosis-... There are more ways to reason than just quantitatively.
167.
by j2kun 9mo ago
Ah yes, let's run a statistical study: give some mentally unstable people ChatGPT and others not, and see if more murder-suicides occur in the treatment group. Oh you mean a correlation study? Well now we can just argue nonstop about r
168.
by j2kun 9mo ago
+1 my point is that, if you didn't learn it, you need a humble attitude. And in the context of software you're building as part of your profession, not learning it well is a bigger problem than hobby pursuits.
169.
by j2kun 9mo ago
> single cases The problem is it's becoming common. How many people have to be convinced by ChatGPT to murder-suicide before you think it's worth doing something?
170.
by j2kun 9mo ago
> I don't know how this doesn't give pause to the ChatGPT team. Especially with their supposed mission to be helpful to the world etc. Because the mission is a lie and the goal is profit. alwayshasbeen.jpg
171.
by j2kun 9mo ago
"Did you learn X before using it?" is indeed my one of my most commonly occurring thoughts that I never say out loud.
172.
by j2kun 10mo ago
Is this something you use to generate static browser tests that no longer use the LLM? Or would you need to use the LLM every time you run the tests?
173.
by j2kun 10mo ago
I had to upgrade to get DDR5, for one.
174.
by j2kun 10mo ago
I'd be shocked if this site was never manipulated by its owners to influence opinion in their benefit.
175.
by j2kun 10mo ago
Read the article, which discusses this already, and maybe respond to that.
176.
by j2kun 10mo ago
It is worth noting that the Netflix Prize winner's solution was never meaningfully used, because Netflix pivoted from ranking content based on what you tell them you like to ranking content based on clicks and minutes watched. To say t
177.
by j2kun 10mo ago
They claim the algorithm "discovered" the new techniques, but the methods described in section 5 do not seem all that novel to me. It smells like it could be "laundering" the literature [1] and reshuffling existing techn
178.
by j2kun 10mo ago
> like building an AI product made me part of the problem. I don't see how the author can believe that quitting their job to work on an AI startup is NOT contributing to the problem of "AI products being shoved down everyone&#x
179.
by j2kun 10mo ago
Scalar Evolution is one way loops can be simplified
180.
by j2kun 10mo ago
...did they exercise it?

More ›