# Can AI learn to give useful scientific feedback? A new paper bets on author compliance as the answer. - Date: 2026-04-15 - Category: Artificial Intelligence The paper's key insight: judge feedback quality by whether the author revised the paper. Train on 19,534 ICLR submissions using that signal, and a small open-weight model beats Gemini-3-flash on precision: not by being smarter, but by learning from what actually worked. ---