Theme of inquiry
What gaps exist between model confidence and actual reasoning quality?
A question within its area, explored through 4 lines of inquiry below — each a family of specific questions the research asks.
83 specific questions
- Can uncertainty estimates based on model self-assessment reliably signal errors?
- Does layer-wise prediction stabilization provide a stronger trace quality signal than confidence alone?
- Can people reliably recognize when an AI is uncertain versus confident?
- Can retrieval density alone correct overreliance on confident but wrong outputs?
- Can intrinsic confidence signals improve both calibration and reasoning performance?
- Can confidence levels reliably detect when a model is overthinking?
- Why do models that repeat errors seem more confident than models that contradict themselves?
119 specific questions
- Do models verbalize their implicit knowledge when that knowledge influences their output?
- Can interpretability tools distinguish genuine reasoning from fabricated reasoning inside models?
- Do internal belief probes reveal what models actually know versus report?
- How misaligned are verbal reports from internal model computation?
- Do models leak their true associations through reasoning traces and behavior?
- Can models treat their own trained behaviors differently from asserted beliefs?
- Can representation engineering reliably identify and manipulate self-referential concepts in models?
81 specific questions
- How does AI fact-checking increase belief in false headlines users saw?
- Can credible fact-checking from political opponents neutralize AI-generated attack content?
- Can users reliably distinguish valid reasoning from plausible-looking deception?
- What makes experience-dependent claims categorically different from other types of fabricated statements?
- Does AI-generated text about personal experiences create a distinct category of falsity?
- Why do users override their own judgment when AI says a headline is false?
- Why does false information spread faster when presupposed rather than asserted?
86 specific questions
- How does self-revision in reasoning chains amplify confidence in wrong answers?
- Why does self-reflection during training fail to improve model self-correction?
- Why does model self-revision increase confidence while degrading accuracy?
- Does self-reflection help models notice their own constraint violations?
- Why do reasoning models struggle with self-evaluation and revision?
- How does self-revision on wrong answers increase model confidence further?
- Why does single-model self-revision amplify confidence in incorrect answers?