The next paradigm: The convoluted maze of abundant model choice, loop systems and the hopeful final solution
The next paradigm: The convoluted maze of abundant model choice, loop systems and the hopeful final solution. I’m going to be talking of things like these: https://youtu.be/BrjAt-wvEXI (Tree of Thoughts - GPT-4 Reasoning is Improved 900% - Wes Roth) https://youtu.be/wVzuvf9D9BU (GPT 4 is Smarter than You Think: Introducing SmartGPT - AI Explained) GPT-4 is not the most accurate model, as some people still think. GPT-4 with this kind of pre-prompting and self-reflection is. I predicted this a while ago. I don’t believe there’s a name for these things yet, but I just call them “loop systems”. These loop systems are the most accurate models that we can possibly use right now. I can only foresee the meta staying this way as well. For people saying that OpenAI should change their base model t...