yusina

8 ms·

61.
by yusina 1y ago
The rule set is simple: "Don't be biased." What does that mean? And that is the problem. It's hard (read: impossible) to define in technical, formal terms. That's because bias is at the root a social problem, not
62.
by yusina 1y ago
Please convince us that these are not just words. As much as I'd want to believe you, the sweet VC money is in the feedback that many people here advise you against. It will be hard to stay away from it.
63.
by yusina 1y ago
> I came to the realization “Code Reviews Are Dead”. If that's how it works at your company then run as fast as you can. There are many reasonable alternatives that won't push this AI-generated BS on you. That is, if you care.
64.
by yusina 1y ago
IMO that's what this focus on. Language. That's what LLMs excel at. Perhaps branch out to providing localized papers to markets like China or France (hah, sorry). Judging the actual contents may feel like the holy grail but is unl
65.
by yusina 1y ago
You don't have much experience in it, do you? The real world peer review process could not be furher from what you are describing. Source: I've personally been involved in peer reviewing in fields as diverse as computer science, q
66.
by yusina 1y ago
> a fair, merit-based AI reviewer That's a dream which is unlikely to come true. One reason being that the training data is not unbiased and it's very hard to make it less biased, let alone unbiased. The other issue being that
67.
by yusina 1y ago
I would say the exact opposite. Let the machine do the easy/simple/boring stuff. Let the human peer reviewer do the big question. That's what the human is good at and excited about and it's what the machine will not be g
68.
by yusina 1y ago
The LLM is only capable of doing so if it has encountered something similar before as part of its training. Claiming anything else requires a proof.
69.
by yusina 1y ago
Sorry, I don't follow. How do you define LLMs?
70.
by yusina 1y ago
You are asking why it is meaningful to use terms for what they mean instead of making up things? Well, I prefer it that way, but the spirit of "AI" seems to go in another direction, and the leadership of US government also does, s
71.
by yusina 1y ago
That's an interesting take and in fact one I could get behind. But I'm afraid that most folks using the term mean it more literally than you describe.
72.
by yusina 1y ago
Extraordinary claim requires extraordinary proof. I don't know, but I'm also not the one claiming something. (Besides, we know what LLMs do, and none of those things indicate understanding. Just statistics.)
73.
by yusina 1y ago
Um, moving the goal post? The claim was LLMs understand things. The counter was, nope, they don't. They can fake it well though. Your argument now is, well humans also often fake it. Kinda implying that it means it's ok to claim t
74.
by yusina 1y ago
That's the bar now? Some tech is great if the average human is worse at it?
75.
by yusina 1y ago
For natural language you are right. The language came first, the grammar was retrofitted to try to find structure. For formal languages, which programming languages (and related ones like query languages, markup languages, etc) are an insta
76.
by yusina 1y ago
> Hacker news people still think LLMs are just some statistical model guessing things. That's exactly what they are. It's the definition of what they are. If you are talking about something that is doing something else, then
77.
by yusina 1y ago
> understand the work LLMs don't understand . It's mind-boggling to me that large parts of the tech industry think that. Don't ascribe to them what they don't have. They are fantastic at faking understanding. Don&#
78.
by yusina 1y ago
> You effectively have an infinite number of agents You don't. Sincerely, Your Electricity Bill
79.
by yusina 1y ago
> So you wrote 'article' but you meant "submission" ;) Ah, true. Sorry.
80.
by yusina 1y ago
> You should know that LLMs are very weak in procedural operations. Indeed! That's why they are LLMs, not AI! Why are new hypes always re-defining terms? 20 years ago, "AI" was actually about intelligence. Just like "
81.
by yusina 1y ago
> You would need to have something like networked driving, or government supported networks of driving information, to deal with the cultural factor. Or actual intelligence. That observes its surroundings and learns what's going o
82.
by yusina 1y ago
> a technology which 3 years ago did not exist Decades of machine learning research would like to have a word.
83.
by yusina 1y ago
This heading is so obvious, I hardly see why it warrants an article. The other day I googled "Is it 2025?" and the AI reply was that nope, it's 2024. Such a joke.
84.
by yusina 1y ago
It's because they are providing abstractions which we/the compilers use, but just doing that would be too slow, so they implement optimizations, but those are based on certain assumptions, so then the users adjust what they do to
85.
by yusina 1y ago
And why do you want to be part of such a group? My approach is: If some group doesn't want me in their midst because we are weird in too different ways, then the feeling is mutual and I move on.
86.
by yusina 1y ago
Exactly, that was my point. Thanks for putting it more succinctly.
87.
by yusina 1y ago
Ok, if that's the class of systems we are talking about, then the system that your decider wants to check does not need to attain the same state twice. It may or may not terminate, and your decider may never know. You can't have i
88.
by yusina 1y ago
And how does your decider recognize that a state has been attained twice? If I make the input system large enough then you don't have enough space in your decider to save all the states which it has observed.
89.
by yusina 1y ago
The halting problem applies also for finite-but-unbounded memory. If you give me a decider that can tell me for any program with a state space up to size N whether it halts or not, then I will be able to produce another program with a large
90.
by yusina 1y ago
I appreciate your lengthy explanation and I largely agree with it, even though it probably doesn't help much because anybody who has not understood this yet will have stopped reading latest at 20% in. That's not your fault but jus

More ›