Self-Evolution

Appears in 1 paper

A bootstrapping process where: (1) a model generates candidate solutions using search, (2) solutions are verified automatically, (3) correct, high-quality solutions become training data, (4) the model is trained on this data, improving for

As used in Paper 24 — rStar-Math: Small LLMs Can Master Math Reasoning with Self-Evolved Deep Thinking →

A bootstrapping process where: (1) a model generates candidate solutions using search, (2) solutions are verified automatically, (3) correct, high-quality solutions become training data, (4) the model is trained on this data, improving for the next round. Each round's improved model generates better solutions for the subsequent round.