1 paper across 1 session
A game theory-based mechanism drives LLMs from self-check to consensus for improved accuracy.
