Login

News on Medial

šŸ”„ 1 person is talking about this

OpenAIā€™s new ā€œCriticGPTā€ model is trained to criticize GPT-4 outputs

ArstechnicaArstechnica Ā· 4m
OpenAIā€™s new ā€œCriticGPTā€ model is trained to criticize GPT-4 outputs

OpenAI has introduced CriticGPT, an AI model designed to detect errors in code generated by ChatGPT. The model aims to improve the alignment of AI systems with human expectations by employing Reinforcement Learning from Human Feedback (RLHF). CriticGPT, based on the GPT-4 family of large language models (LLMs), assists human reviewers by identifying coding mistakes in the output. The researchers trained CriticGPT on a dataset of intentionally bugged code to teach it how to recognize and flag different errors. According to experiments, CriticGPT's critiques were preferred by trainers over human critiques in 63% of cases involving LLM errors.

12
2
View Source

Comments

More Like this

OpenAIā€™s ā€˜year of the enterpriseā€™ includes new tools for increasing AI accuracy

OpenAIā€™s ā€˜year of the enterpriseā€™ includes new tools for increasing AI accuracy

OpenAI makes ChatGPT 'more direct, less verbose'

OpenAI makes ChatGPT 'more direct, less verbose'

OpenAI Wants AI to Help Humans Train AI

OpenAI Wants AI to Help Humans Train AI

ETtech Explainer: Is Appleā€™s ReALM better than OpenAIā€™s GPT-4?

ETtech Explainer: Is Appleā€™s ReALM better than OpenAIā€™s GPT-4?

OpenAI introduces CriticGPT, a GPT 4-based model that can detect errors in ChatGPTā€™s code output

OpenAI introduces CriticGPT, a GPT 4-based model that can detect errors in ChatGPTā€™s code output

1960s chatbot ELIZA beat OpenAIā€™s GPT-3.5 in a recent Turing test study

1960s chatbot ELIZA beat OpenAIā€™s GPT-3.5 in a recent Turing test study

OpenAIā€™s Sora Turns AI Prompts Into Photorealistic Videos

OpenAIā€™s Sora Turns AI Prompts Into Photorealistic Videos

This AI stock trader engaged in insider trading ā€” despite being instructed not to ā€“ and lied about it

This AI stock trader engaged in insider trading ā€” despite being instructed not to ā€“ and lied about it

1960s chatbot ELIZA beat OpenAIā€™s GPT-3.5 in a recent Turing test study

1960s chatbot ELIZA beat OpenAIā€™s GPT-3.5 in a recent Turing test study

In an ocean of white data, this founder is trying to make AI less biased. Just don't call it 'BlackGPT.'

In an ocean of white data, this founder is trying to make AI less biased. Just don't call it 'BlackGPT.'

Download the medial app to read full posts, comements and news.