Calibration Toolkit for Generative Models

Amélie Roussel

doi:10.xxxx/ctm.ar.2026.0007

Research ArchiveSoftwareVersion 0.4.0 Open access

Calibration Toolkit for Generative Models

Amélie Roussel · Department of Computer Science, Meridian University

Published 5 June 2026

DOI 10.xxxx/ctm.ar.2026.0007 · ID CTM-AR-2026-0007

Download files

Sample record. Fictional demonstration content; the files listed are not attached.

Description

Python library for measuring and improving calibration of generated text, including reliability diagrams, expected calibration error variants and post-hoc recalibration methods.

Keywords calibration · software · uncertainty

Files (1)

Name	Format	Size	Download
calibration-toolkit-0.4.0.zip	ZIP	1.1 MB	Download

Files are checksummed on upload. Demonstration record: no files are attached.

Related works

Versions

Version 0.4.0 · 5 June 2026Latest
Per-retriever calibration plots.
10.xxxx/ctm.ar.2026.0007.v0.4.0
Version 0.3.0 · 4 May 2026
First public release.
10.xxxx/ctm.ar.2026.0007.v0.3.0

How to cite

Roussel, A. (2026). Calibration Toolkit for Generative Models (Version 0.4.0) [Software]. CONTINUUM Research Archive. doi:10.xxxx/ctm.ar.2026.0007