From Answers to Interpretations: Rethinking Ambiguity-Induced Aleatoric Uncertainty Estimation in LLMs
This work argues that answers are not necessary for identifying ambiguity: they are often redundant, add avoidable cost, and can mislead through epistemic leakage, and suggests that ambiguity-induced aleatoric uncertainty is better estimated from the interpretation space than from the response space.
Omer Nahum, Niv Nayman, Jonathan Fhima et al.
· 0 citations