Olga Lavinda
Titles as the sources gave them. In 1 entry, with 0 checked quotations.
In the Base
Every entry they appear in.
Known Good, Known Bad: A Ground-Truth Calibration Protocol for AI Judgment in Science
According to the abstract, Olga Lavinda proposes a protocol that tests AI judgment against a verified reference case and deviant variants, and demonstrates it with an organic chemistry example in which a free ChatGPT model detected three of five intended deviations at first.