Gold News reported that OpenAI stated that it has trained a model based on GPT-4, named CriticGPT, to capture errors in the output of ChatGPT code. It is beginning to integrate models similar to CriticGPT into our RLHF to provide clear artificial intelligence assistance for OpenAI trainers. The plan is to further expand the application scale of RLHF on GPT-4 and put it into practice.
Voir l'original
Cette page peut inclure du contenu de tiers fourni à des fins d'information uniquement. Gate ne garantit ni l'exactitude ni la validité de ces contenus, n’endosse pas les opinions exprimées, et ne fournit aucun conseil financier ou professionnel à travers ces informations. Voir la section Avertissement pour plus de détails.
OpenAI lance CriticGPT, un modèle pour détecter les erreurs de code de ChatGPT
Gold News reported that OpenAI stated that it has trained a model based on GPT-4, named CriticGPT, to capture errors in the output of ChatGPT code. It is beginning to integrate models similar to CriticGPT into our RLHF to provide clear artificial intelligence assistance for OpenAI trainers. The plan is to further expand the application scale of RLHF on GPT-4 and put it into practice.