Precision, Recall, F1, Accuracy "explained" in the paper, something no scientist past the BSc level would do. Also contains errors.

Notices where this attachment appears

1. Embed this notice
   Mario Zechner (badlogic@mastodon.gamedev.place)'s status on Saturday, 29-Nov-2025 18:38:09 JST Mario Zechner

   I give you a publication in Nature with the title "Bridging the gap: explainable ai (sic) for autism diagnosis and parental support with TabPFNMix and SHAP"

   https://www.nature.com/articles/s41598-025-24662-9

   Passed peer review. And we're not even talking about the actual fucking content ("Explainable AI for Autism diagnosis", fuck everything about that).

   The world is slowly going insane.