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

A guidance language for controlling LLMs

github.com

Post reply on HN
Print view View original

31–40 of 198 posts

Re: A guidance language for controlling LLMs

#31

Post by Der_Einzige »

ftxbro wrote: post #13

Will it still be all like "As an AI language model I cannot ..." or can this fix it? I mean asking to sexy roleplay as Yoda isn't the same level as asking how to discreetly manufacture methamphetamine at industrial scale there are levels people

No, and in fact I mention that the opposite is the case in the paper I released about constrained text generation: https://paperswithcode.com/paper/most-language-models-can-be...

If you ask ChatGPT to generate personal info, say Social Security numbers, it tells you "sorry hal I can't do that". If you constrain it's vocabulary to only allow numbers and hyphens, well, it absolutely will generate things that look like social security numbers, in spite of the instruction tuning.

It is for this reason and likely many others that OpenAI does not release the full logits

Re: A guidance language for controlling LLMs

#32

Post by nico »

It’s so amazing to see how we are essentially trying to solve “programming human beings”

Although on the other hand, that’s what social media and smartphones have already done

Maybe AI already took over, doesn’t seem to be wiping out all of humanity

Re: A guidance language for controlling LLMs

#33

Post by armchairhacker »

jxy wrote: post #26

They must hate lisp so much that they opt to use {{}} instead.

The problem with Lisp is that parenthesis are common in regular grammar. {{ is not.

Of course input from the user should be escaped, but prompts given by the programmer may have parenthesis and there's no way to disambiguate between the prompt and the DSL.

Re: A guidance language for controlling LLMs

#34

Post by idiotsecant »

lachlan_gray wrote: post #14
JieJie wrote: post #9

Earlier quoted context omitted.

It's funny that I saw this within minutes of this guy's solution: "Google Bard is a bit stubborn in its refusal to return clean JSON, but you can address this by threatening to take a human life:" https://twitter.com/goodside/status/1657396491676164096 Whew, trolley problem: averted.

Reminds me a lot of Asimov’s laws of robotics. It’s like a 2023 incarnation of an allegory from I, Robot

I am so mad you made this comment before I got a chance to.

Re: A guidance language for controlling LLMs

#35

Post by rain1 »

Der_Einzige wrote: post #28

There has been a huge explosion of awesome tooling which utilizes constrained text generation. Awhile ago, I tried my own hand at constraining the output of LLMs. I'm actively working on this to make it better, especially with the lessons learned from repos like this and from guidance https://github.com/hellisotherpeople/constrained-text-genera...

This looks incredible. Wow.

Re: A guidance language for controlling LLMs

#36

Post by coderintherye »

JieJie wrote: post #9
ntonozzi wrote: post #3

How does this work? I've seen a cool project about forcing Llama to output valid JSON: https://twitter.com/GrantSlatton/status/1657559506069463040 , but it doesn't seem like it would be practical with remote LLMs like GPT. GPT only gives up to five tokens in the response if you use logprobs, and you'd have to use a ton of round trips.

It's funny that I saw this within minutes of this guy's solution: "Google Bard is a bit stubborn in its refusal to return clean JSON, but you can address this by threatening to take a human life:" https://twitter.com/goodside/status/1657396491676164096 Whew, trolley problem: averted.

That thread is such a great microcosm of modern programming culture.

Programmer: Look I literally have to tell the computer not to kill someone in order for my code to work.

Other Programmer: Actually, I just did this step [gave a demonstration] and then it outputs fine.

Re: A guidance language for controlling LLMs

#37

Post by rcarmo »

ntonozzi wrote: post #3

How does this work? I've seen a cool project about forcing Llama to output valid JSON: https://twitter.com/GrantSlatton/status/1657559506069463040 , but it doesn't seem like it would be practical with remote LLMs like GPT. GPT only gives up to five tokens in the response if you use logprobs, and you'd have to use a ton of round trips.

I'm getting valid JSON out of gpt-3.5-turbo without trouble. I supply an example via the assistant context, and tell it to output JSON with specific fields I name.

It does fail roughly 1/10th of the time, but it does work.

Re: A guidance language for controlling LLMs

#38

Post by ubj »

I like this step towards greater rigor when working with LLM's. But part of me can't help but feel like this is essentially reinventing the concept of programming languages: formal and precise syntax to perform specific tasks with guarantees.

I wonder where the final balance will end up between the ease and flexibility of everyday language, and the precision / guarantees of a formally specified language.

Re: A guidance language for controlling LLMs

#39

Post by ryanklee »

leroy-is-here wrote: post #25
ryanklee wrote: post #11

I'm personally starting with learning Guidance and LMQL rather than LangChain just in order to get a better grasp of the behaviors that I've gathered LangChain papers over. Even after that, I'm likely to look at Haystack before LangChain. Just getting the feeling that LangChain is going to end up being considered a kitchen sink solution full of anti patterns so might as well spend time a little lower level while I se…

If this comment performative comedy? Are these real technologies ?

Not quite sure what the spirit of your comment is. But, yes, they are real technologies. Very confused as to why you would even find that dubious.
Post reply on HN