Reinforcement Learning with Metacognitive Feedback Elicits Faithful Uncertainty Expression in LLMs

Part of

Context Sensitivity In Generative AI