Corpus KAS 2.0: Cleaner and with New Datasets

Corpus of Academic Slovene (KAS) contains Slovene BSc/BA, MSc/MA, and PhD theses from 2000 - 2018. We present a cleaner version of the corpus with added text segmentation and updated POS-tagging. The updated corpus of abstracts contains fewer artefacts. Using machine learning classifiers, we filled in miss- ing research field information in the metadata. We used the full texts and corresponding abstracts to create several new datasets: monolingual and cross-lingual datasets for long text summariza- tion of academic texts and a dataset of aligned sentences from abstracts in English and Slovene, suitable for machine transla- tion. We release the corpora, datasets, and developed source code under a permissible licence