{"type":"link","version":"1.0","title":"Feeding a model its own highest-likelihood output as memory adds no information, however close to the reference it is","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/risk-likelihood-picked-self-output-adds-no-information","thumbnail_url":"https://ai-arch.pages.dev/og/risk-likelihood-picked-self-output-adds-no-information.png","thumbnail_width":1200,"thumbnail_height":630}