{"type": "article", "title": "Abstention as an Action Can Kill Both the Reward Gradient and the KL Anchor: Collapse Law and Repair for Error-Penalized Reinforcement Learning", "publisher": "Web Pulse", "url": "https://wpnews.pro/news/abstention-as-an-action-can-kill-both-the-reward-gradient-and-the-kl-anchor-law", "original_source": "https://arxiv.org/abs/2608.00301", "published": "2026-08-04T04:00:00+00:00", "accessed": "2026-08-04", "id": "abstention-as-an-action-can-kill-both-the-reward-gradient-and-the-kl-anchor-law"}