{"type":"link","version":"1.0","title":"A policy-learned memory-selection step outperforms semantic top-k retrieval at reconstructing bounded agent context","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/learned-selection-beats-semantic-retrieval-for-context-reconstruction","thumbnail_url":"https://ai-arch.pages.dev/og/learned-selection-beats-semantic-retrieval-for-context-reconstruction.png","thumbnail_width":1200,"thumbnail_height":630}