3 ms·
At that point why not just have LLMs generate bytecode in one shot? Plenty of training data to go on, I'd imagine.
by senbrow 1y ago
At that point why not just have LLMs generate bytecode in one shot?
Plenty of training data to go on, I'd imagine.
- dyauspitr 1y agoThe code would be un reviewable.
- TeMPOraL 1y agoIt would also be harder for the LLM to work with. Much like with humans, the model's ability to understand and create code is deeply intertwined and inseparable from its general NLP ability.
- senbrow 1y agoWhy couldn't you use an LLM to generate source code from a prompt, compile it, then train a new LLM on the same prompt using the compiled output? It seems no different in kind to me than image or audio generation.
- senbrow 1y ago...by a human :)
- dyauspitr 1y agoHence very important in the transitional phase we are currently in where LLMs can’t do everything yet.