{"type":"link","version":"1.0","title":"Benchmark-passing agent PRs have a 24.2 percentage-point lower human merge rate than benchmark scores predict","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/benchmark-passing-overpredicts-production-merge","thumbnail_url":"https://ai-arch.pages.dev/android-chrome-512x512.png","thumbnail_width":512,"thumbnail_height":512}