no code implementations • GWC 2019 • Jan Kocoń, Arkadiusz Janz
In this paper we present a novel method for emotive propagation in a wordnet based on a large emotive seed.
no code implementations • GWC 2019 • Francis Bond, Arkadiusz Janz, Marek Maziarz, Ewa Rudnicka
According to George K. Zipf, more frequent words have more senses.
no code implementations • GWC 2019 • Francis Bond, Arkadiusz Janz, Maciej Piasecki
In this paper, we compare a variety of sense-tagged sentiment resources, including SentiWordNet, ML-Senticon, plWordNet emo and the NTU Multilingual Corpus.
no code implementations • EACL (GWC) 2021 • Arkadiusz Janz, Marek Maziarz
We propose a novel method of homonymy-polysemy discrimination for three Indo-European Languages (English, Spanish and Polish).
no code implementations • EACL (GWC) 2021 • Arkadiusz Janz, Maciej Piasecki, Piotr Wątorski
Neural language models, including transformer-based models, that are pre-trained on very large corpora became a common way to represent text in various tasks, including recognition of textual semantic relations, e. g. Cross-document Structure Theory.
no code implementations • GWC 2018 • Maciej Piasecki, Gabriela Czachor, Arkadiusz Janz, Dominik Kaszewski, Paweł Kędzia
The paper presents construction of large scale test datasets for word embeddings on the basis of a very large wordnet.
no code implementations • GWC 2018 • Gabriela Czachor, Maciej Piasecki, Arkadiusz Janz
Word embeddings were used for the extraction of hyponymy relation in several approaches, but also it was recently shown that they should not work, in fact.
no code implementations • GWC 2018 • Jan Kocoń, Arkadiusz Janz, Maciej Piasecki
In this paper we present a comprehensive overview of recent methods of the sentiment propagation in a wordnet.
no code implementations • 14 Feb 2024 • Stanisław Woźniak, Bartłomiej Koptyra, Arkadiusz Janz, Przemysław Kazienko, Jan Kocoń
Large language models (LLMs) have significantly advanced Natural Language Processing (NLP) tasks in recent years.
no code implementations • 31 May 2023 • Konrad Wojtasik, Vadim Shishkin, Kacper Wołowiec, Arkadiusz Janz, Maciej Piasecki
In this work, inspired by mMARCO and Mr.~TyDi datasets, we translated all accessible open IR datasets into Polish, and we introduced the BEIR-PL benchmark -- a new benchmark which comprises 13 datasets, facilitating further development, training and evaluation of modern Polish language models for IR tasks.
1 code implementation • 21 Feb 2023 • Jan Kocoń, Igor Cichecki, Oliwier Kaszyca, Mateusz Kochanek, Dominika Szydło, Joanna Baran, Julita Bielaniewicz, Marcin Gruza, Arkadiusz Janz, Kamil Kanclerz, Anna Kocoń, Bartłomiej Koptyra, Wiktoria Mieleszczenko-Kowszewicz, Piotr Miłkowski, Marcin Oleksy, Maciej Piasecki, Łukasz Radliński, Konrad Wojtasik, Stanisław Woźniak, Przemysław Kazienko
Our comparison of its results with available State-of-the-Art (SOTA) solutions showed that the average loss in quality of the ChatGPT model was about 25% for zero-shot and few-shot evaluation.
1 code implementation • 23 Nov 2022 • Łukasz Augustyniak, Kamil Tagowski, Albert Sawczyn, Denis Janiak, Roman Bartusiak, Adrian Szymczak, Marcin Wątroba, Arkadiusz Janz, Piotr Szymański, Mikołaj Morzy, Tomasz Kajdanowicz, Maciej Piasecki
In this paper, we introduce LEPISZCZE (the Polish word for glew, the Middle English predecessor of glue), a new, comprehensive benchmark for Polish NLP with a large variety of tasks and high-quality operationalization of the benchmark.
no code implementations • LREC 2020 • Arkadiusz Janz, {\L}ukasz Kopoci{\'n}ski, Maciej Piasecki, Agnieszka Pluwak
Relation Extraction is a fundamental NLP task.
no code implementations • RANLP 2019 • Arkadiusz Janz, Maciej Piasecki
Word Sense Disambiguation remains a challenging NLP task.
no code implementations • RANLP 2017 • Pawe{\l} K{\k{e}}dzia, Maciej Piasecki, Arkadiusz Janz
Similarity between sentences is calculated from graph, and the similarity values are input to classifiers trained by Logistic Model Tree.