News for hackers. Stuff that matters.

A comment on Recursive Language Models. Parent comment

Re:Recursive Language Models by AlexCoventry on Saturday January 03, 2026 @03:50PM (#46481392)
No, there's no training going on, here, as far as I can tell. E.g., they use GPT-5 as their base model. Also, AFAICT from a quick skim/search there's no mention of loss functions or derivatives, FWIW.
[ Parent | on HN ]
Re:Recursive Language Models by alextheparrot on Saturday January 03, 2026 @08:57PM (#46484001)
The derivative being a grad(ient) student sampling scaffolds against evals + qualitative observations: most prompt-based llm papers
[ Parent | on HN ]