Line of inquiry
Inquiring lines›Where does language-model reasonin…›How do reward models guide reliabl…›this line of inquiry
How should models express uncertainty rather than forced confident answers?
A broader line of inquiry — a family of 35 specific questions the research asks around this. Follow one into its inquiring-line page, or move sideways to a related line below.
Questions in this line of inquiry 35
Specific inquiring lines the field asks around this — ordered from the most general framing down to the most specific angle.
- How does expressing uncertainty help models avoid the answer-or-abstain dilemma?
- Do base models and reasoning models fail in opposite directions on uncertainty?
- Why do models report commitment instead of truth uncertainty?
- Does uncertainty quantification in model responses reduce persuasive impact on audiences?
- When models lack representation depth, does refusal look identical to safety-driven over-abstention?
- How does uncertainty estimation drive computational resource allocation in models?
- Why do models maintain accurate beliefs but generate false claims?
- Does distillation strip away uncertainty signals that reasoning actually needs?
- Can models distinguish between logical impossibility and their own execution limits?
- Why do linguistic hedging markers correlate with internal confidence signals in reasoning traces?
- How do unstated feasibility constraints affect model decision-making?
- Why do models commit to answers early on easy versus hard tasks?
- Why does self-distillation suppress epistemic verbalization in student models?
- Can machine learning encode pragmatic reasoning about when rules should bend?
- How can models select the optimal question to ask given multiple uncertainties?
- How does uncertainty verbalization change student robustness across domains?
- What makes uncertainty calibration harder than expanding knowledge?
- Can we measure indifference to truth separately from hallucination rates?
- How do models decide between refusing or hallucinating?
- What makes correcting a false assumption harder than just detecting it?
- Can architectural changes reorder when uncertainty and empowerment signals influence decisions?
- How do one-sided explanations act as confidence signals to users?
- Do models learn different sophistry strategies for QA versus code generation?
- What cognitive structures do realistic belief models need to include?
- Why do weak belief tracking and conservative actions trap agents in low-information states?
- Can we measure sophistry by tracking conviction density in model outputs?
- What makes a first answer so often the best answer a model produces?
- Can agents escape weak belief tracking and conservative action selection traps?
- What distinctive properties make open foundation models different from closed ones?
- How do moment-to-moment ToM fluctuations shape AI response quality?
- How does uncritical acceptance of information relate to silent agreement failures?
- Can belief networks from interviews simulate how people change their minds?
- What mechanism causes confident false answers under high cognitive load?
- Why should low-probability severe risks trigger early intervention?
- How does Peircean Secondness differ from what RLHF actually provides?