Confidence Calibration and Rationalization for LLMs via Multi-Agent Deliberation

Open in new window