Polish Discourse Corpus ISO 24617-8 dataset for DISRPT 2025

This dataset contains the re-annotation of a subset of the Polish Coreference Corpus (PCC) with ISO 24617-8 discourse relations. This work is part of the Universal Discourse project (2023/50/A/HS2/00559) carried out at the Institute of Computer Science of the Polish Academy of Sciences, funded by the National Science Center (NCN). The dataset was used in the DISRPT 2025 shared task and is also available on GitHub: https://github.com/disrpt/sharedtask2025/tree/master/data/pol.iso.pdc