跳到正文
原文
OpenAI News·· 2022-01-27精选AI 评分79

OpenAI 发布 InstructGPT 并将其设为 API 默认模型

Aligning language models to follow instructions

AI 导读

OpenAI 推出 InstructGPT 模型,该模型在遵循用户意图、真实性和减少毒性方面优于 GPT-3,现已作为默认语言模型部署于 API。这些模型采用人类反馈强化学习技术训练,源自 OpenAI 的对齐研究成果。

推荐理由

OpenAI 官方宣布 InstructGPT 成为 API 默认模型,标志着其通过人类反馈强化学习提升指令遵循能力并降低毒性的技术正式落地。

原文

We’ve trained language models that are much better at following user intentions than GPT-3 while also making them more truthful and less toxic, using techniques developed through our alignment research. These InstructGPT models, which are trained with humans in the loop, are now deployed as the default language models on our API.

来源:OpenAI News · openai.com