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

Chain of Recursive Thoughts: Make AI think harder by making it argue with itself

github.com

Post reply on HN
Print view View original

221–230 of 253 posts

Re: Chain of Recursive Thoughts: Make AI think harder by making it argue with itself

#221

Post by voidspark »

casenmgreen wrote: post #210
consumer451 wrote: post #191

Earlier quoted context omitted.

To push back on this, a somewhat recent Linus Torvalds ~quote: "I don't think that 'just predicting the next word' is the insult that people think it is, it's mostly what we all do." If we break our lives down into the different types of reasoning, and what we mostly do day-to-day, this rings very true to me. I currently believe that our brains generally operate as very efficient inference machines. Sometimes we slow…

I may be wrong, but it seems to me this also is a case of improper use of words. Those LLMs neither agree nor disagree. They do not understand. They produce output , and we read that output and we ourselves consider the output to be something, or something else. All an LLM does is produce output. There's no conceptual understanding behind it, and so there is no agreement, or disagreement.

LLMs learn high-dimensional representations that capture conceptual relationships in their training data. They manipulate those representations in ways that approximate human reasoning.

Re: Chain of Recursive Thoughts: Make AI think harder by making it argue with itself

#222

Post by voidspark »

dttze wrote: post #111
consumer451 wrote: post #13

Earlier quoted context omitted.

I am not sure that you can make that absolute statement. Reasoning is subdivided into types, and one of those types is inductive reasoning. > Inductive reasoning refers to a variety of methods of reasoning in which the conclusion of an argument is supported not with deductive certainty, but with some degree of probability. Unlike deductive reasoning (such as mathematical induction), where the conclusion is certain, g…

Markov chains have done that for ages. They aren't AI. This is just that scaled up. Just because it can infer a token doesn't mean it can infer a conclusion to an argument.

> This is just that scaled up

An LLM is not a Markov process. They are fundamentally different. An LLM conditions the next token prediction on the entire context window (via the attention mechanism), not just the previous token. Besides the token history window it also maintains a cache of neural activations which is updated at every step.

Otherwise you could use the same reasoning to argue that a human is a Markov process, which is absurd, but vacuously true if "state" means the quantum level configuration of every atom in the body.

Re: Chain of Recursive Thoughts: Make AI think harder by making it argue with itself

#223

Post by schnitzelstoat »

I probably don't understand the modern, complex models. But doesn't it basically predict the next token given the context and the better models use more training data and can consider a larger context, and have more parameters to better retain information from the training data etc.

But the fundamental way they operate is the same - predicting the next token given previous tokens. Where/how does reasoning happen here?

Re: Chain of Recursive Thoughts: Make AI think harder by making it argue with itself

#224

Post by kmacdough »

Lerc wrote: post #54

I kind of want to try something like this at a larger scale in an always-on mode where I have a 'senate' of debate. Rather than responding to prompts on a case by case basis, provide a list of tasks (potentially with deadlines) and let the senate work on them, break off into groups to manage subtasks, challenge results , make suggestions. Even potentially a tree of analysts where suggestions only gets passed up the t…

These ensembles have been tested throughout AI progress. Well scaffolded larger models have historically come out ahead in both quality and speed/cost.

Perhaps this is a parricularly effective ensemble, but I would need to see real data.

Re: Chain of Recursive Thoughts: Make AI think harder by making it argue with itself

#225

Post by macrolime »

nojs wrote: post #198
hu3 wrote: post #192

Earlier quoted context omitted.

> ...so you need a checker that actually reasons about the world (compiler, linter, SAT solver, ground-truth dataset, etc.). Agree. What do you think about telling the LLM to also generate unit tests for the code it spits and then run all tests (including previous application unit tests). I think this is a way to ensure some level of grounded verification: - Does code compile? - Do unit test pass? AI can then consume…

This works well but only if you eyeball the tests and edit them a bit in my experience. Otherwise it gets lazy and makes them trivial to pass. Also, you’ve often gotta explicitly tell it not to hardcode test cases in the solution to make them pass.

Yet, having the LLM evaluate the tests often works well. Especially if you ask it to look for hardcoded test cases.

Re: Chain of Recursive Thoughts: Make AI think harder by making it argue with itself

#226

Post by captainbland »

keepamovin wrote: post #211
dudeinhawaii wrote: post #190

I see a lot of threads pitting models against each other (or whole swarms of them) in the hope that "wisdom of crowds" will magically appear. After a stack of experiments of my own—and after watching the recent ASU/Microsoft-Research work [1].. I've landed on a simpler takeaway: An LLM is a terrible verifier of another LLM. Subbarao Kambhampati's "(How) Do LLMs Reason/Plan?" talk shows GPT-4 confidently producing pro…

I believe, what the smart AI company is trying to do, right now, in secret, is to use US, the humans, and our replies to the AIs, as training for the next generation of self-verifying-models. :) Training on corpus data gets you to 1 order of magnitude. But training on interactive data where you can observe and adapt to the OODA-loop? So much more powerful. At least, that's what I'd be doing if I were doing AI :) But…

I think you'd need to screen for quality of response quite stringently as loads of people will produce "corrections" which are just plain wrong.

Re: Chain of Recursive Thoughts: Make AI think harder by making it argue with itself

#228

Post by vintermann »

bee_rider wrote: post #131
vunderba wrote: post #91

Earlier quoted context omitted.

A year or so ago I experimented with splitting a user prompt down to a set of "different AI personas" that would each try to approach the user's problem in a different way and then bubble back up with a master arbiter for consensus. I modeled it after the concept of advisors from Civilization II. It worked reasonably well though I think it was at least somewhat limited by being constrained to a single LLM (Mistral).…

What sort of personalities did you try? A group where some members have grudges against each other and will irrationally poke holes in each other’s plans could be a fun experiment.

I have played Disco Elysium, and can confirm that a bunch of inner voices arguing with each other can be fun.

Re: Chain of Recursive Thoughts: Make AI think harder by making it argue with itself

#229

Post by small_scombrus »

schnitzelstoat wrote: post #223

I probably don't understand the modern, complex models. But doesn't it basically predict the next token given the context and the better models use more training data and can consider a larger context, and have more parameters to better retain information from the training data etc. But the fundamental way they operate is the same - predicting the next token given previous tokens. Where/how does reasoning happen here…

Upfront: I think AI is borderline useless for many of the tasks we give it. But:

1. Do our neurons not just react the same way every time to the same input? A brain is larger than the sum of its parts.

2. They don't reason, but you can somewhat emulate (or pretend to be) reasoning if you feed something back into itself enough times and it pinky promises reasoning is happening

Re: Chain of Recursive Thoughts: Make AI think harder by making it argue with itself

#230

Post by BOOSTERHIDROGEN »

nativeit wrote: post #174
BOOSTERHIDROGEN wrote: post #172

Earlier quoted context omitted.

How?

Ask the AI assistant for instructions. Pretty soon we'll have new acronyms such as "IDKATFAIA" ["I don't know, ask the f'ing AI already"] as we all succumb to the knowledge soup.

Honestly, the AI assistant isn't as smart as I thought - I'm still having to check its work.
Post reply on HN