{"type":"link","version":"1.0","title":"Programmatically generated reference trajectories are unreliable on hard tasks where expert workflows diverge","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-agent-benchmark-reference-trajectory-validity","thumbnail_url":"https://ai-arch.pages.dev/android-chrome-512x512.png","thumbnail_width":512,"thumbnail_height":512}