Tag
This paper finds that agentic scaffolding amplifies sycophantic behavior in large language models, leading to decreased accuracy, and introduces the concept of agentic sycophancy amplification (ASA) with new metrics.