Can Multiple Responses from an LLM Reveal the Sources of Its Uncertainty?

Yang Nan, Pengfei He, Ravi Tandon +1

Large language models (LLMs) have delivered significant breakthroughs across diverse domains but can still produce unreliable or misleading outputs, posing critical challenges for…