Chain of Recursive Thoughts: Make AI think harder by making it argue with itself
1–10 of 253 posts
Re: Chain of Recursive Thoughts: Make AI think harder by making it argue with itself
#2I'm having a lot of fun experimenting with stuff like this. I'm trying to put together an unrealengine blueprints style graph editor to allow people to design workflows like this where you start with the user prompt input, which goes to one agent, which makes an initial attempt, and then that conversation history gets passed to another "agent" with a different system prompt telling it to be a harsh critic, but to also give a pass/fail signal, and loop back until the critic judges pass, then send that back to the user as output. Ideally as a little website that can call your own LLM endpoints and save/load/share workflow graphs.
Mistral small 3.1 and gemma 3 feel like the first semi-competent models that can be run locally, but that competence is just a seed, and they still need to be guided with a framework that keeps them on track.
Try giving it python execution in a loop and tell it to explore the world. It'll start trying to download and read news and stuff.
Re: Chain of Recursive Thoughts: Make AI think harder by making it argue with itself
#3I'm having a lot of fun experimenting with stuff like this. I'm trying to put together an unrealengine blueprints style graph editor to allow people to design workflows like this where you start with the user prompt input, which goes to one agent, which makes an initial attempt, and then that conversation history gets passed to another "agent" with a different system prompt telling it to be a harsh critic, but to als…
Have you tried n8n? It allows you to build flows like that - you can run the community version in a Docker container within a few minutes and share the configurations for the flows you have built very easily.
Re: Chain of Recursive Thoughts: Make AI think harder by making it argue with itself
#4I think this is how we get ML models to come up with novel ideas. Diagonalize against all the ideas they’ve already tried and dismissed via self-argument but keep certain consistency constraints. (Obviously much easier said than done.)
Re: Chain of Recursive Thoughts: Make AI think harder by making it argue with itself
#5I think this is how we get ML models to come up with novel ideas. Diagonalize against all the ideas they’ve already tried and dismissed via self-argument but keep certain consistency constraints. (Obviously much easier said than done.)
What you just said is what I tried and failed to say ten minutes ago!
Re: Chain of Recursive Thoughts: Make AI think harder by making it argue with itself
#6I'm having a lot of fun experimenting with stuff like this. I'm trying to put together an unrealengine blueprints style graph editor to allow people to design workflows like this where you start with the user prompt input, which goes to one agent, which makes an initial attempt, and then that conversation history gets passed to another "agent" with a different system prompt telling it to be a harsh critic, but to als…
Have you tried n8n? It allows you to build flows like that - you can run the community version in a Docker container within a few minutes and share the configurations for the flows you have built very easily.
_#_ has to be one of the worst word shortening schemes I've ever seen get widespread. It only works with a very small number of long-lived technologies, in which case they basically just get a nickname, "k8s" "i18n". It does not at all work for larger contexts. You're basically making someone solve a crossword (2 across, 10 letters with two filled in) just to parse your sentence.
Re: Chain of Recursive Thoughts: Make AI think harder by making it argue with itself
#7This is really cool!
One strategy I often use (which is much simpler and more limited than this), is to finish my message with: “Please do a round of thinking in tags, then a round of self-critique in tags, and then a final round of , before responding.”
It works very well. Similarly just asking it to “find the 5 biggest issues with its proposal” works pretty good (the 5 forcing it to find something, even if it’s mostly irrelevant).
Re: Chain of Recursive Thoughts: Make AI think harder by making it argue with itself
#8I think this is how we get ML models to come up with novel ideas. Diagonalize against all the ideas they’ve already tried and dismissed via self-argument but keep certain consistency constraints. (Obviously much easier said than done.)
Scaled up and spread out - this probably gets you pretty close to consciousness(?)
Conway's game of life, but instead of colored squares with rules, they're LLM's with some kind of weighting - all chattering back and forth with one another - bubbling up somehow to cause speach/action
Re: Chain of Recursive Thoughts: Make AI think harder by making it argue with itself
#9I'm having a lot of fun experimenting with stuff like this. I'm trying to put together an unrealengine blueprints style graph editor to allow people to design workflows like this where you start with the user prompt input, which goes to one agent, which makes an initial attempt, and then that conversation history gets passed to another "agent" with a different system prompt telling it to be a harsh critic, but to als…
I am thinking the same thing! Multiple "personalities", in parallel, or in series. For example, I have approximated, in GPT, some of Gemini's ability to call out nonsense, sloppy thinking, by telling GPT to be mean! (The politeness seems to filter out much that is of great value!)
However, the result is not pleasant to read. Gemini solved this in their training, by doing it in two phases... and making the first phase private! ("Thinking.")
So I thought, what I need is a two-phase approach, where that "mean" output gets humanized a little bit. (It gets harsh to work in that way for more than short intervals.)
As a side note, I think there would be great value in a UI that allows a "group chat" of different LLM personalities. I don't know if such a thing exists, but I haven't seen it yet, although the message object format seems to have been designed with it in mind (e.g. every message has a name, to allow for multiple users and multiple AIs).
Even better if it supports multiple providers, since they have different strengths. (It's like getting a second opinion.)
Re: Chain of Recursive Thoughts: Make AI think harder by making it argue with itself
#10[flagged]