Results

You can review the results through our evaluation portal but here are the medallists:

Task 1: Punctuation restoration from read text
  1. Krzysztof Wróbel (eNeLPol / UJ / AGH) – 81.29
  2. Norbert Ropiak (CLARIN) – 81.25
  3. Michał Marcińczuk (Samurai Labs / Wrocław University of Science and Technology) – 81.23
Task 2a: Evaluation of translation quality assessment metrics (nonblind)
  1. Darek Kłeczek (skok.ai) – 0.6137
  2. Artur Nowakowski (Adam Mickiewicz University) – 0.5728
  3. Krzysztof Wróbel (eNeLPol / UJ / AGH) – 0.5677
Task 2b: Evaluation of translation quality assessment metrics (blind)
  1. Krzysztof Wróbel (eNeLPol / UJ / AGH) – 0.5138
  2. Darek Kłeczek (skok.ai) – 0.4840
  3. Artur Nowakowski (Adam Mickiewicz University) – 0.4793
Task 3: Post-correction of OCR results
  1. Mateusz Piotrowski – 3.744
  2. Krzysztof Wróbel (eNeLPol / UJ / AGH) – 4.302
  3. Paweł Dyda (Adam Mickiewicz University / Applica.ai) – 5.328
Task 4: Question answering challenge
  1. Mateusz Piotrowski – 71.68
  2. Aleksander Smywiński–Pohl (AGH / eNeLPol) – 50.96
    Piotr Rybak (ML Research at Allegro.pl) – 50.96
  3. Darek Kłeczek (skok.ai) – 46.44