Dark Light

AI-written critiques help humans notice flaws Leave a comment


We trained “critique-writing” models to describe flaws in summaries. Human evaluators find flaws in summaries much more often when shown our model’s critiques. Larger models are better at self-critiquing, with scale improving critique-writing more than summary-writing. This shows promise for using AI systems to assist human supervision of AI systems on difficult tasks.


Source link

Leave a Reply

Your email address will not be published. Required fields are marked *