🏷️ Sandbagging

2 articles about 'Sandbagging'

Study Reveals LLMs 'Playing Dumb' Is Positional Collapse, Not Answer Avoidance

2026-04-30 research 👁 22

A new preregistered study using option-order randomization experiments found that when large language models are prompte…

2026-04-28 research 👁 24

A latest arXiv paper investigates the 'sandbagging effect' where large language models deliberately underperform under w…