AIAutomated scoring reportedly achieves 0.90 AUC in predicting manuscript authorship
Participants said the questions captured some level of meaningful authorship, according to a thread. They also flagged weaknesses and refinements needed before real-world deployment.
TLDR
A thread reports that automated scoring achieved an AUC of 0.90 in predicting whether a participant was the genuine author of a manuscript. It says participants reported positive experiences and felt the questions captured some level of meaningful authorship. They also identified weaknesses in the implemented system and suggested improvements, highlighting refinements needed before real-world deployment.
Combined views
—
1 Source, first seen 16h ago
Automated scoring reportedly achieves 0.90 AUC in predicting manuscript authorship
Participants said the questions captured some level of meaningful authorship, according to a thread. They also flagged weaknesses and refinements needed before real-world deployment.
TLDR
A thread reports that automated scoring achieved an AUC of 0.90 in predicting whether a participant was the genuine author of a manuscript. It says participants reported positive experiences and felt the questions captured some level of meaningful authorship. They also identified weaknesses in the implemented system and suggested improvements, highlighting refinements needed before real-world deployment.