Ask HN: Is reasoning just some iterative context refinement?

2 pointssschmitt1 comment

context = "you are a useful model. you evaluate your own output for N steps after the initial user input"

prompt = "here comes a prompt by the user"

context += prompt

for _ in range(N): context += evaluate_llm(context)

Are you asking in terms of how the current crop of "reasoning models" are implemented, or are you asking more philosophically about the nature of true reasoning?

Calling it "refinement" is dismissive. It's generating new information, which is in many cases well beyond the scope of the original prompt.

Reasoning models today are just a marketing spin on chain-of-thought techniques that benefit from reinforcement learning.

Reply on news.ycombinator.com