Either this already exists, or someone is going to implement that (should I implement that?): - assumption LLM can input/output in any useful language, - human languages are not exactly optimal away to talk with LLM, - internally LLMs keep knowledge as whole bunch of connections with some weights and multiple layers, - they need to decode human-language input into tokens, then into something that is easy to digest by further layers, then get some output, translate back into tokens and human language (or programming language, same thing), - this whole human language <-> tokens <-> input <-> LLM <-> output <-> tokens <-> language is quite expensive.
What if we started to talk to LLMs in non-human readable languages (programming languages are also just human readable)? Have a tiny model run locally that translates human input, code, files etc into some-LLM-understandable-language, LLM gets this as an input, skips bunch of layers in input/output, returns back this non-human readable language, local LLM translates back into human language/code changes.
Yesterday or two days ago there was a post about using Apple Fundamental Models, they have really tiny context window. But I think it could be used as this translation layer human->LLM, LLM->human to talk with big models. Though initially those LLMs need to discover which is "language" they want to talk with, feels like doable with reinforcement learning. So cheap local LLM to talk to big remote LLM.
Either this is done already, or it's a super fun project to do.
My theory was that someone should write a specific LLM language, and then spend a whole lot of money to train models using that. A few times other commenters here have pointed out that that would be really difficult .
But I think you're onto something, human languages just aren't optimal here. But to actually see this product to conclusion you'd probably need 60 to 100 million. You would have to completely invent a new language and awesome invent new training methods on top of it.
I'm down if someone wants to raise a VC round.
I'm currently downloading Ollama and going to write a simple proof-of-concept with Qwen as local "frontend", talking to OpenAI GPT as "backend". I think the idea is sound, but indeed needs retraining of GPT (hmm like training tiny local LLM in synchronization of a big remote LLM). It might be not bad business venture in the end.
I don't think humans should be involved in developing this AI-AI language, just giving some guidance, but let two agents collaborate to invent the language, and just gratify/punish them with RL methods.
OpenAI looking at you, got an email some days ago "you're not using OpenAI API that much recently, what changed?"
If you want to start a Git repo somewhere let me know and I'll do what I can to help.
I imagine it's possible, but just a manner of money.