Smoothed Elicitation Complexity for Approximate Γ \Gamma -calibration of Discrete Classification Tasks

Jessica Finocchiaro, Victor Ganson, Drona Khurana · 2026

Read Smoothed Elicitation Complexity for Approximate Γ \Gamma -calibration of Discrete Classification Tasks by Jessica Finocchiaro, Victor Ganson, Drona Khurana (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-23017.txt.

Description

Finocchiaro, Ganson, and Khurana study rough calibration for discrete traits of outcome distributions, such as the mode or rankings. Binary calibration extends poorly to multiclass settings, where prediction spaces grow exponentially with class count. Noarov and Roth eased that blowup by judging traits of the distribution instead, measured by elicitation complexity. Prior work covered mostly continuous scalar traits. The authors use Lipschitz continuous traits as go-betweens to handle discrete traits that are strongly orderable, and give algorithms for designing them.

Catalog record

CREATORJessica Finocchiaro, Victor Ganson, Drona Khurana
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE47883
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2605.23017

Browse the full library · About the Tower of Babel