Skip to content
Publications

Book · Zenodo · October 2, 2026

Trustworthy confidence for Kazakh morphological tagging codebase

Laura M. Baitenova, Akhmed Baikhojayev, Г.С. Мухамеджанова, Gauhar Munaitbas, Сакен Мамбетов, Daryn Nurtaza

Abstract

Code and measured results for the paper "Trustworthy Confidence for Kazakh Morphological Tagging: Calibration and Selective Decision-Making in a Low-Resource Text-Correction Pipeline". One code path measures four taggers — KazBERT, KazRoBERTa, XLM-R base and a feature-based CRF — on the canonical 754/162/162 split of UD_Kazakh-KTB (7,376 / 1,566 / 1,594 tokens, seed 42, decontaminated), with temperature fitted on dev and every calibration/selective-prediction metric evaluated on the held-out test split.

Natural Language Processing TechniquesTopic ModelingSentiment Analysis and Opinion Miningtemperature scalingcalibrationselective prediction

More by Laura Baitenova