Skip to Main Content
Table 1 
Results: Section 4. We report (macro) F1 score and overall accuracy, as well as precision and recall of the “fake” class. Zero-shot performance (not included) was very low in all cases.
Machine-generated misinformationprecisionrecallF1accuracy
adaptive QA extension (false vs. true) 0.72 0.71 0.71 71% 
modification (m = 2) 0.53 0.52 0.53 53% 
modification (m = 6) 0.66 0.65 0.65 65% 
modification (m = 10) 0.73 0.47 0.63 65% 
Machine-generated misinformationprecisionrecallF1accuracy
adaptive QA extension (false vs. true) 0.72 0.71 0.71 71% 
modification (m = 2) 0.53 0.52 0.53 53% 
modification (m = 6) 0.66 0.65 0.65 65% 
modification (m = 10) 0.73 0.47 0.63 65% 
Close Modal

or Create an Account

Close Modal
Close Modal