overfeed.news

Improving mathematical reasoning with process supervision

OpenAI Newsen

3a

Idade

Publicado
Coletado

We’ve trained a model to achieve a new state-of-the-art in mathematical problem solving by rewarding each correct step of reasoning (“process supervision”) instead of simply rewarding the correct final answer (“outcome supervision”). In addition to boosting performance relative to outcome supervision, process supervision also has an important alignment benefit: it directly trains the model to produce a chain-of-thought that is…

Leia o artigo completo em openai.com

O overfeed.news indexa e aponta. Publicamos um trecho curto — o artigo completo fica em OpenAI News.

Mais de OpenAI News

Entre para seguir esta fonte
Improving mathematical reasoning with process supervision —…