{"type":"link","version":"1.0","title":"Weight distillation candidates by how useful each one is downstream, rather than keeping only the best or averaging over all","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/weight-distillation-candidates-by-downstream-usefulness","thumbnail_url":"https://ai-arch.pages.dev/og/weight-distillation-candidates-by-downstream-usefulness.png","thumbnail_width":1200,"thumbnail_height":630}