This paper describes our submission for the 4th Shared Task on SlavNER on three Slavic languages - Czech, Polish and Russian. We use pre-trained multilingual XLM-R Language Model (Conneau et al., 2020) and fine-tune it for three Slavic languages using datasets provided by organizers. Our multilingual NER model achieves 0.896 F-score on all corpora, with the best result for Czech (0.914) and the worst for Russian (0.880). Our cross-language entity linking module achieves F-score of 0.669 in the official SlavNER 2023 evaluation.

Although information on the Internet can be shared in many languages, the language presence on the World Wide Web is very disproportionate. The problem of multilingualism on the Web, in particular access, availability and quality of information in the world’s languages, has been the subject of UNESCO focus for several decades. Making European websites more multilingual is also one of the focal targets of the Connecting Europe Facility Automated Translation (CEF AT) digital service infrastructure. In order to monitor this goal, alongside other possible solutions, CEF AT needs a methodology and easy to use tool to assess the degree of multilingualism of a given website. In this paper we investigate methods and tools that automatically analyse the language diversity of the Web and propose indicators and methodology on how to measure the multilingualism of European websites. We also introduce a prototype tool based on open-source software that helps to assess multilingualism of the Web and can be independently run at set intervals. We also present initial results obtained with our tool that allows us to conclude that multilingualism on the Web is still a problem not only at the world level, but also at the European and regional level.

Named entity recognition, in particular for morphological rich languages, is challenging task due to the richness of inflected forms and ambiguity. This challenge is being addressed by SlavNER Shared Task. In this paper we describe system submitted to this task. Our system uses pre-trained multilingual BERT Language Model and is fine-tuned for six Slavic languages of this task on texts distributed by organizers. In our experiments this multilingual NER model achieved 96 F1 score on in-domain data and an F1 score of 83 on out of domain data. Entity coreference module achieved F1 score of 47.6 as evaluated by bsnlp2021 organizers.