Check out “Recursive Language Models”, or RLMs.

I believe this method works well because it turns a long context problem (hard for LLMs) into a coding and reasoning problem (much better!). You’re leveraging the last 18 months of coding RL by changing you scaffold.

▲

koakuma-chan 7 hours ago | parent [-]

This seems really weird to me. Isn't that just using LLMs in a specific way? Why come up with a new name "RLM" instead of saying "LLM"? Nothing changes about the model.

▲

vimda 7 hours ago | parent [-]

RLMs are a new architecture, but you can mimic an RLM by providing the context through a tool, yes

	▲	anonymousd3vil 4 hours ago \| parent [-]
		New architecture to building agent, but not the model itself. You still have LLMs, but you kinda give this new agentic loop with a REPL environment where the LLM can try to solve the problem more programmatically.