{"type":"link","version":"1.0","title":"Self-rewarding training loops inflate response length round after round, confounding judge-based quality gains","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-self-rewarding-loops-inflate-response-length","thumbnail_url":"https://ai-arch.pages.dev/og/risk-self-rewarding-loops-inflate-response-length.png","thumbnail_width":1200,"thumbnail_height":630}