AI-written critiques assist people discover flaws

We educated “critique-writing” fashions to explain flaws in summaries. Human evaluators discover flaws in summaries way more usually when proven our mannequin’s critiques. Bigger fashions are higher at self-critiquing, with scale enhancing critique-writing greater than summary-writing. This exhibits promise for utilizing AI methods to help human supervision of AI methods on troublesome duties.

Evolution by giant fashions

Strategies for coaching giant neural networks