§ feed · storyline

Summarizing books with human feedback

OpenAI publishes research on using human feedback to train AI systems to summarise books, scaling oversight to tasks that are hard for humans to directly evaluate.

Sep 23 · 09:00:00 · primary fetch1 sourceupdated Sep 23 · 09:00:00

Scaling human oversight of AI systems for tasks that are difficult to evaluate.

read full article on openai.com ↗

§ sources1 publication · timeline below

openai.comSummarizing books with human feedbackprimary09:00:00