We trained “critique-writing” models to explain flaws in summaries. Human evaluators find flaws in summaries way more often when shown our model’s critiques. Larger models are higher at self-critiquing, with scale improving critique-writing greater...
We recognize that identifying AI-written text has been a very important point of debate amongst educators, and equally necessary is recognizing the boundaries and impacts of AI generated text classifiers within the classroom. We've...
We’re launching a classifier trained to differentiate between AI-written and human-written text.
We’ve trained a classifier to differentiate between text written by a human and text written...