Theme of inquiry

Why does model confidence diverge from actual reasoning quality?

A question within its area, explored through 7 lines of inquiry below — each a family of specific questions the research asks.


Can LLMs genuinely introspect or only simulate self-awareness?

39 specific questions

See all 39 questions in this line of inquiry
What's the relationship between persuasiveness and factual accuracy?

28 specific questions

See all 28 questions in this line of inquiry
Why does self-revision fail to improve and instead amplify confidence?

54 specific questions

See all 54 questions in this line of inquiry
How do confident outputs distort user judgment of accuracy?

30 specific questions

See all 30 questions in this line of inquiry
What internal signals best predict whether reasoning will succeed?

57 specific questions

See all 57 questions in this line of inquiry
Why does voting over multiple reasoning samples improve model performance?

28 specific questions

See all 28 questions in this line of inquiry
How does training for improved reasoning reduce abstention ability?

44 specific questions

See all 44 questions in this line of inquiry