Exploration of the Equation Space Utilizing RAG Based LLM’s in a Opposition Format.
In this short summary we go over how AI agents can be used in equation farming.
The idea is simple. If a LLM can now perform research at 50% accuracy at a advanced level, then it will get some new powerful things right. Thus the cycle is simple. Ask the LLM to invent a new theory that solves a currently unknown problem set, then it opposes itself to try to prove it's own work wrong. If the new theory fails, repeat the process.
- This burns a LOT of tokens, it is suggested to use a non-per-token based subscription.
By this method we can conduct 'equation farming' the wholesale streamline processing of theoretical research.
