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

131–140 of 253 posts

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

#131

Post by bee_rider »

vunderba wrote: post #91
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…

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.

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

#132

Post by mountainriver »

electroly wrote: post #22

This seems to be different than I expected from the title. I thought it would be explicitly adversarial. 1. You are the assistant. Please answer the question directly. 2. You are the cross-examiner. The assistant is wrong. Explain why. 3. You are the assistant. The cross-examiner is wrong. Defend your claim. 4. You are a judge. Did either party make their case, or is another round of argumentation required? I haven't…

Techniques like this have been around since GPT-3.5. There are boatloads of papers on the topic.

I have no idea why anyone thinks this is novel. I guess that speaks to the state of HN

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

#135

Post by moribunda »

mountainriver wrote: post #132
electroly wrote: post #22

This seems to be different than I expected from the title. I thought it would be explicitly adversarial. 1. You are the assistant. Please answer the question directly. 2. You are the cross-examiner. The assistant is wrong. Explain why. 3. You are the assistant. The cross-examiner is wrong. Defend your claim. 4. You are a judge. Did either party make their case, or is another round of argumentation required? I haven't…

Techniques like this have been around since GPT-3.5. There are boatloads of papers on the topic. I have no idea why anyone thinks this is novel. I guess that speaks to the state of HN

Exactly... I thought that implementing STORM was just a basic step in this topic... Looks like we're running in circles.

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

#138

Post by itissid »

odo1242 wrote: post #88

Something I do sometimes is: - Have an AI chat model come up with an answer to a problem. - Have it write a report discussing the details of the problem and why it's answer is correct, directed at a person or AI model who has no knowledge of the initial problem or technical field. - Have a second AI model with no knowledge of the problem grade the report, and write it's own report either (a) asking for clarification…

Isn't this kind of another way of how Inference Time Scaling works? It will basically produce several chain of thoughts and then pursue one that has maximum reward based on an internal function?

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

#139

Post by stevenAthompson »

dttze wrote: post #119
stevenAthompson wrote: post #20

Earlier quoted context omitted.

Can you define "thinking" in a way that excludes what the AI is doing, but includes what humans do? I haven't' really seen anyone else manage it without talking about ghosts or some other kind of metaphysical voodoo.

Using a conceptual understanding of something to deduce or infer something else. An LLM doesn't know what anything is. Just what goes around the token representation of that thing.

"conceptual understanding of something" is just another way of saying "the relationship between concepts", which is exactly what transformer models use.

*EDIT* To elaborate, how can you define anything in isolation of every other concept/thing? You can't. Things are only defined by their relationships to each other, which is exactly the same thing transformer models do.

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

#140

Post by _ea1k »

odo1242 wrote: post #88

Something I do sometimes is: - Have an AI chat model come up with an answer to a problem. - Have it write a report discussing the details of the problem and why it's answer is correct, directed at a person or AI model who has no knowledge of the initial problem or technical field. - Have a second AI model with no knowledge of the problem grade the report, and write it's own report either (a) asking for clarification…

This reminds me a lot of the YT video that went over using Monte Carlo Tree Search with LLMs to maximize result quality. Link: https://www.youtube.com/watch?v=mfAV_bigdRA&ab_channel=Treli...

It seemed like a pretty good idea, though I'd guess that it would greatly increase token usage. I'd also be concerned that the LLM as a judge might struggle to grade things accurately if it wasn't also able to generate good enough answers to begin with.

Post reply on HN