OpenAI ↗ · 2023-05-31 15:00 · UTC+8
Improving mathematical reasoning with process supervision
公司动态 OpenAI 原文语言:英文
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…
原文摘录
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…
以下是来源提供的简短摘录,完整内容请查看原文。