Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

LLMs as the new high level language

federicopereiro.com

Post reply on HN
Print view View original

211–220 of 396 posts

Re: LLMs as the new high level language

#211

Post by avaer »

anoplus wrote: post #138
eastbound wrote: post #132

Earlier quoted context omitted.

There’s barely any debate, people don’t answer each other; It’s rather about invoking the wonder and imagination of everyone’s brain. Like spatial conquest or an economic crisis: It will change everything but you can’t do anything immediately about it, and everyone tries to understand what it will change so they can adapt. It’s more akin to 24hrs junk news cycle, where everything is presented as an alert but your tem…

I am interested in a post that will teach me how to consume only the news that really relevant for me

The hidden fallacy in your comment is that there is such a thing as "news that is really relevant for you".

This isn't all that different than saying that it would be nice if someone else did your thinking for you -- which is a totally fine thing to want, but let's not get confused.

"News that is relevant for you" is a concept made up by advertising companies to legitimize them in having power over what you see. Because if they presented it plainly, you would be rightly alarmed.

Re: LLMs as the new high level language

#212

Post by fauigerzigerk »

red75prime wrote: post #195
zekica wrote: post #147

Earlier quoted context omitted.

You can't think all the way about refining your prompt for LLMs as they are probabilistic. Your re-prompts are just retrying until you hit a jackpot - refining only works to increase the chance to get what you want. When making them deterministic (setting the temperature to 0), LLMs (even new ones) get stuck in loops for longer streams of output tokens. The only way to make sure you get the same output twice is to us…

Randomness is not a problem by itself. Algorithms in BQP are probabilistic too. Different prompts might have different probabilities of successful generation, so refinement could be possible even for stochastic generation. And provably correct one-shot program synthesis based on an unrestricted natural language prompt is obviously an oxymoron. So, it's not like we are clearly missing the target here.

>Different prompts might have different probabilities of successful generation, so refinement could be possible even for stochastic generation.

Yes, but that requires a formal specification of what counts as "success".

In my view, LLM based programming has to become more structured. There has to be a clear distinction between the human written specification and the LLM generated code.

If LLMs are a high level programming language, it has to be clear what the source code is and what the object code is.

Re: LLMs as the new high level language

#213

Post by BudapestMemora »

matheus-rr wrote: post #210

The intermediate product argument is the strongest point in this thread. When we went from assembly to C, the debugging experience changed fundamentally. When we went from C to Java, how we thought about memory changed. With LLMs, I'm still debugging the same TypeScript and Python I was before. The generation step changed. The maintenance step didn't. And most codebases spend 90% of their life in maintenance mode. Th…

"Until prompt-to-binary is reliable enough that nobody reads the intermediate code, the analogy doesn't hold."

1. OK, let's create 100 instances of prompt under the hood, 1-2 will hallucinate, 3-5 will produce something different from 90% of remaining, and it can compile based on 90% of answers

2. computer memory is also not 100% reliable , but we live with it somehow without man-in-the-middle manually check layer?

Re: LLMs as the new high level language

#217

Post by Lwerewolf »

BudapestMemora wrote: post #213
matheus-rr wrote: post #210

The intermediate product argument is the strongest point in this thread. When we went from assembly to C, the debugging experience changed fundamentally. When we went from C to Java, how we thought about memory changed. With LLMs, I'm still debugging the same TypeScript and Python I was before. The generation step changed. The maintenance step didn't. And most codebases spend 90% of their life in maintenance mode. Th…

"Until prompt-to-binary is reliable enough that nobody reads the intermediate code, the analogy doesn't hold." 1. OK, let's create 100 instances of prompt under the hood, 1-2 will hallucinate, 3-5 will produce something different from 90% of remaining, and it can compile based on 90% of answers 2. computer memory is also not 100% reliable , but we live with it somehow without man-in-the-middle manually check layer?

I wonder what ECC is for. So, unless you're Google and you're having to deal with "mercurial cores"...

Also, sorry, but what did I just actually attempt to read?

Re: LLMs as the new high level language

#218

Post by BudapestMemora »

And sooner or later it will happen, imho. With probabalistic compiling. And several "prompts/agents" under the hood. The majority of "replies" wins to compile. Of course good context will contribute to better refined probability.

Ask yourself "Computer memory and disk are also not 100% reliable , but we live with it somehow without man-in-the-middle manual check layer, yes?" Answer about LLM will be the same, if good enough level of similarity/same asnwers is achieved.

Re: LLMs as the new high level language

#219

Post by booleandilemma »

hndc wrote: post #95
ChrisGreenHeur wrote: post #89

Earlier quoted context omitted.

I may have bad news for you on how compilers typically work.

Deterministic compilation, aka reproducible builds, has been a basic software engineering concept and goal for 40+ years. Perhaps you could provide some examples of compilers that produce non-deterministic output along with your bad news.

Account created 11 months ago. They're probably just some slop artist with too much confidence. They probably don't even know what a compiler is.

Re: LLMs as the new high level language

#220

Post by fulafel »

bandrami wrote: post #203
fulafel wrote: post #200

Earlier quoted context omitted.

Programs written in traditional PLs are also often probabilistic. It seems that the same mechanisms could be used to address this in both types (formal methods).

Huh? What's an example of a probabilistic programming language?

Race conditions, effects of memory safety and other integrity bugs, behaviours of distributed systems, etc.
Post reply on HN