{"type":"link","version":"1.0","title":"RoMeRL: Balancing Feedback Coverage and the Memory-Reward Trap in Self-Evolving Agent Memory via Reduced-Order Utility States","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/arxiv2608.02508","thumbnail_url":"https://ai-arch.pages.dev/og/arxiv2608.02508.png","thumbnail_width":1200,"thumbnail_height":630}