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

201–210 of 253 posts

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

#201

Post by visarga »

the_af wrote: post #127
nonethewiser wrote: post #62

Earlier quoted context omitted.

Chatgpt shares context between chats. I wonder how that impacts it? It seems like a good approach though. What you dont want to do is ever suggest that its wrong yourself. Usually it will just assume it is wrong. Actually what I find impressive is when I do this and it actually pushes back to defend itself.

Does it share context even if no "memory updated" message appears indicating it has stored a fact about you? I asked ChatGPT and it says no, but then again it's not reliable at introspection or at revealing data about how it works.

I think they are different systems, one is a collection of saved snippets and the other more like RAG over chat history.

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

#202

Post by aprilthird2021 »

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 takes such a long time to do though, no? What problems does this save you time on?

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

#204

Post by StopDisinfo910 »

meander_water wrote: post #162
StopDisinfo910 wrote: post #126

Earlier quoted context omitted.

For anything semi-adversarial, I have had good results asking the AI to come up with a plan, then take the side of the opponent coming with counter play/way to defeat the plan, finally asking for a revision of the initial plan given the potential reaction from the opponent. The final plan you obtain is generally a lot more well rounded and thought out. I find that amusing because the technique also works when I apply…

To be honest, this is what I assumed this repo was doing from the title. It talks about arguing with itself, but it looks like it's just generating multiple alternative responses in parallel and selecting the best one. Do you find your method handles "sycophancy" well?

I don’t really know.

I stopped using ChatGPT at some point because I disliked how cagey it became about a lot of topics. I used to enjoy making write improbable movies mashup when GPT3 was released and at some point it became very touchy about IP rights and violence which was annoying.

I generally use Deepseek nowadays which is not sycophantic and surprisingly doesn’t seem as censored to me especially if you use a version not hosted by Deepseek themselves.

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

#205

Post by caseyy »

zamalek wrote: post #158
caseyy wrote: post #79

I tried something similar when Llama2 came out, pitting two assistants, who each believed the other is the user, against each other. Ultimately, it was the same model talking with itself. The system prompts for both had various instructions to disagree and criticise the opinion of the user. I provided the first message to get things started. Usually, it’s be along the lines of “nuclear proliferation is harmful to hum…

This might be a situation that warrants a higher temperature. Actually, it could be worth starting a very high temperature initially and gradually decreasing it.

Even after turning the temperature way up, the outcome was the same, just the text less coherent. Not dismissing the idea, just sharing my exp.

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

#206

Post by kgeist »

albertgoeswoof wrote: post #125

How far is this going to go? Are we going to have a team of AI agents that runs a scrum team and meets for stand ups every couple of hours? Are we going to replicate government bureaucracy with agents all debating topics all day long to find the best opinion?

I once attended a talk a year ago where a techlead did just that - they had AI agents that ran a scrum team with different roles, each agent's prompt was to disagree with everyone else (or be highly critical) and present their own point of view, and then an arbiter would make the final decision. They claimed it worked for them.

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

#207

Post by swyx »

meander_water wrote: post #199
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…

For better or worse this has become the defacto standard in LLM Evaluation research papers since the LLM as a Judge paper [0] came out. Its also heavily embedded into frameworks like LangChain and LlamaIndex to evaluate RAG pipelines. [0] https://arxiv.org/abs/2306.05685 [1] https://arxiv.org/abs/2411.15594

its for the better, and i'm actually serious about this. it's just that Subbarao is ALSO right and it is not perfect nor human level. but it -DOES- improve results measurably and consistently.

so what i'm saying is don't throw the baby out with the bathwater. LLM as judge doesnt replace human judgement but its a pretty darn good first pass for how cheap it is. and you can imagine that it will get better over time.

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

#208

Post by asdfman123 »

ivape wrote: post #150
asdfman123 wrote: post #133

And when I do this people say I'm overanalyzing

The thing that makes us weird to regular people is what's going to make us uniquely positioned to utilize AI. If people only knew the level at which I overanalyze and entertain weird ideas. I always inject these personality quirks into my instructions and get very creative results. In a weird way, I'm starting to appreciate just how weird I actually am.

I don't actually think it's that weird though

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

#209

Post by looofooo0 »

_ea1k wrote: post #140
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 enoug…

If you think about marginal cost, such experiments can be run almost at only the cost of extra electricity used for that computation, which in Europe is often zero, at least by the ones who own the compute.

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

#210

Post by casenmgreen »

consumer451 wrote: post #191
casenmgreen wrote: post #116

Earlier quoted context omitted.

To add a bit to this : expert systems have two properties. They give an answer, and they explain their reasoning . LLM cannot explain their reasoning, and that is because there is no reasoning.

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.

Post reply on HN