Abstention as an Action Can Kill Both the Reward Gradient and the KL Anchor: Collapse Law and Repair for Error-Penalized Reinforcement Learning

Xujun Che, Yuchen Yuan, Weida Zhao +1

Error-penalized scoring rules ($+1$ for a correct answer, $-\hat{I}»$ for a wrong one, $0$ for abstaining) are increasingly prescribed against hallucination: a rational agent facing such…