no code implementations • 30 Nov 2023 • Matúš Pikuliak, Andrea Hrckova, Stefan Oresko, Marián Šimko
We present GEST -- a new dataset for measuring gender-stereotypical reasoning in masked LMs and English-to-X machine translation systems.
1 code implementation • 15 Nov 2023 • Ivan Vykopal, Matúš Pikuliak, Ivan Srba, Robert Moro, Dominik Macko, Maria Bielikova
Automated disinformation generation is often listed as an important risk associated with large language models (LLMs).
1 code implementation • 20 Oct 2023 • Dominik Macko, Robert Moro, Adaku Uchendu, Jason Samuel Lucas, Michiharu Yamashita, Matúš Pikuliak, Ivan Srba, Thai Le, Dongwon Lee, Jakub Simko, Maria Bielikova
There is a lack of research into capabilities of recent LLMs to generate convincing text in languages other than English and into performance of detectors of machine-generated text in multilingual settings.
1 code implementation • 13 May 2023 • Matúš Pikuliak, Ivan Srba, Robert Moro, Timo Hromadka, Timotej Smolen, Martin Melisek, Ivan Vykopal, Jakub Simko, Juraj Podrouzek, Maria Bielikova
Fact-checkers are often hampered by the sheer amount of online content that needs to be fact-checked.
1 code implementation • 24 Feb 2023 • Matúš Pikuliak, Ivana Beňová, Viktor Bachratý
Many measures of societal bias in language models have been proposed in recent years.
1 code implementation • 3 Jan 2023 • Matúš Pikuliak, Marián Šimko
This position paper discusses the problem of multilingual evaluation.
1 code implementation • 30 Sep 2021 • Matúš Pikuliak, Štefan Grivalský, Martin Konôpka, Miroslav Blšták, Martin Tamajka, Viktor Bachratý, Marián Šimko, Pavol Balážik, Michal Trnka, Filip Uhlárik
We introduce a new Slovak masked language model called SlovakBERT.
no code implementations • 18 Sep 2018 • Andrej Švec, Matúš Pikuliak, Marián Šimko, Mária Bieliková
First, we detect inappropriate comments for moderators to see.