{"type":"link","version":"1.0","title":"LLM agents given raw state numbers without stated changes barely react, and the simulation goes unrealistically flat","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-llm-simulation-without-explicit-change-cues-goes-flat","thumbnail_url":"https://ai-arch.pages.dev/og/risk-llm-simulation-without-explicit-change-cues-goes-flat.png","thumbnail_width":1200,"thumbnail_height":630}