{"type":"link","version":"1.0","title":"Taxing an optional reasoning step by its cost is an alternative to isolating its reward credit for preventing shortcut behavior","author_name":"AI Archs","author_url":"https://ai-arch.pages.dev","provider_name":"AI Archs","provider_url":"https://ai-arch.pages.dev","url":"https://ai-arch.pages.dev/n/cost-gated-reward-shaping-substitutes-for-isolating-credit","thumbnail_url":"https://ai-arch.pages.dev/android-chrome-512x512.png","thumbnail_width":512,"thumbnail_height":512}