Gold standard corpora (GSCs) are essential for the supervised training and evaluation of systems that perform natural language processing (NLP) tasks. Currently, most of the resources used in biomedical NLP tasks are mainly in English. Little effort has been reported for other languages including Romanian and, thus, access to such language resources is poor. In this paper, we present the construction of the first morphologically and terminologically annotated biomedical corpus of the Romanian language (MoNERo), meant to serve as a gold standard for biomedical part-of-speech (POS) tagging and biomedical named entity recognition (bioNER). It contains 14,012 tokens distributed in three medical subdomains: cardiology, diabetes and endocrinology, extracted from books, journals and blogposts. In order to automatically annotate the corpus with POS tags, we used a Romanian tag set which has 715 labels, while diseases, anatomy, procedures and chemicals and drugs labels were manually annotated for bioNER with a Cohen Kappa coefficient of 92.8% and revealed the occurrence of 1877 medical named entities. The automatic annotation of the corpus has been manually checked. The corpus is publicly available and can be used to facilitate the development of NLP algorithms for the Romanian language.

Transformer models produce advanced text representations that have been used to break through the hard challenge of natural language understanding. Using the Transformer’s attention mechanism, which acts as a language learning memory, trained on tens of billions of words, a word sense disambiguation (WSD) algorithm can now construct a more faithful vectorial representation of the context of a word to be disambiguated. Working with a set of 34 lemmas of nouns, verbs, adjectives and adverbs selected from the National Reference Corpus of Romanian (CoRoLa), we show that using BERT’s attention heads at all hidden layers, we can devise contextual vectors of the target lemma that produce better clusters of lemma’s senses than the ones obtained with standard BERT embeddings. If we automatically translate the Romanian example sentences of the target lemma into English, we show that we can reliably infer the number of senses with which the target lemma appears in the CoRoLa. We also describe an unsupervised WSD algorithm that, using a Romanian BERT model and a few example sentences of the target lemma’s senses, can label the Romanian induced sense clusters with the appropriate sense labels, with an average accuracy of 64%.

LegalNERo is a manually annotated corpus for named entity recognition in the Romanian legal domain. It provides gold annotations for organizations, locations, persons, time and legal resources mentioned in legal documents. Additionally it offers GEONAMES codes for the named entities annotated as location (where a link could be established). The LegalNERo corpus is available in different formats: span-based, token-based and RDF. The Linguistic Linked Open Data (LLOD) version is provided in RDF-Turtle format. CONLLUP files conform to the CoNLL-U Plus format https://universaldependencies.org/ext-format.html . Part-of-speech tagging was realized using UDPIPE. Named entity annotations are placed in the column "RELATE:NE" (the 11th column) as defined in the "global.columns" metadata field. Similarly GEONAMES references are in the column "RELATE:GEONAMES" (the 12th column, last). Automatic processing was performed through the RELATE platform (https://relate.racai.ro). ANN files conform to BRAT format (https://brat.nlplab.org/). The archive contains: - ann_LEGAL_PER_LOC_ORG_TIME_overlap Folder in which all the files are in .ann format and contains annotations of: legal resources mentioned, persons, locations, organizations and time. Overlapping annotations of organizations and time entities inside legal references were allowed. - ann_LEGAL_PER_LOC_ORG_TIME Folder in which all the files are in .ann format and contains annotations of: legal resources mentioned, persons, locations, organizations and time. Overlapping annotations were not allowed and only the longest named entities were annotated. - ann_PER_LOC_ORG_TIME Folder in which all the files are in .ann format and contains annotations of: persons, locations, organizations and time. There are no overlapping annotations. - conllup_LEGAL_PER_LOC_ORG_TIME Folder in which all the files are in .conllup format and contains annotations of: legal resources mentioned, persons, locations, organizations and time. Overlapping annotations were not allowed and only the longest named entities were annotated. The annotation of these files was enhanced with GEONAMES codes (where linking was possible). - conllup_PER_LOC_ORG_TIME Folder in which all the files are in .conllup format and contains annotations of: persons, locations, organizations and time. Overlapping annotations were not allowed and only the longest named entities were annotated. The annotation of these files was enhanced with GEONAMES codes (where linking was possible). - rdf Folder containing the corpus in RDF-Turtle format. All the annotations are available here in both span and token format. - text Folder containing the raw texts. LICENSING This work is provided under the license CC BY-NC-ND 4.0 (Attribution-NonCommercial-NoDerivatives 4.0 International). The license can be viewed online here: https://creativecommons.org/licenses/by-nc-nd/4.0/ and the full text here: https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode . CONTACT Research Institute for Artificial Intelligence "Mihai Draganescu", Romanian Academy Web: http://www.racai.ro Contact emails: vasile@racai.ro , maria@racai.ro

In recent years, research into multimodal resources, models, and tools has seen significant advancements, primarily due to the development of deep artificial neural network architectures. Even though large foundation models exist for different multimodal tasks, their utilization is hindered by a lack of resources for certain languages. In this paper, we explore the existing multimodal resources available for the Romanian language while identifying the current challenges and perspectives in this area. We aim to foster further research into multimodal resources and tools focusing on the Romanian language.

LegalNERo is a manually annotated corpus for named entity recognition in the Romanian legal domain. It provides gold annotations for organizations, locations, persons, time and legal resources mentioned in legal documents. Additionally it offers GEONAMES codes for the named entities annotated as location (where a link could be established). The LegalNERo corpus is available in different formats: span-based, token-based and RDF. The Linguistic Linked Open Data (LLOD) version is provided in RDF-Turtle format. CONLLUP files conform to the CoNLL-U Plus format https://universaldependencies.org/ext-format.html . Part-of-speech tagging was realized using UDPIPE. Named entity annotations are placed in the column "RELATE:NE" (the 11th column) as defined in the "global.columns" metadata field. Similarly GEONAMES references are in the column "RELATE:GEONAMES" (the 12th column, last). Automatic processing was performed through the RELATE platform (https://relate.racai.ro). ANN files conform to BRAT format (https://brat.nlplab.org/). The archive contains: - ann_LEGAL_PER_LOC_ORG_TIME_overlap Folder in which all the files are in .ann format and contains annotations of: legal resources mentioned, persons, locations, organizations and time. Overlapping annotations of organizations and time entities inside legal references were allowed. - ann_LEGAL_PER_LOC_ORG_TIME Folder in which all the files are in .ann format and contains annotations of: legal resources mentioned, persons, locations, organizations and time. Overlapping annotations were not allowed and only the longest named entities were annotated. - ann_PER_LOC_ORG_TIME Folder in which all the files are in .ann format and contains annotations of: persons, locations, organizations and time. There are no overlapping annotations. - conllup_LEGAL_PER_LOC_ORG_TIME Folder in which all the files are in .conllup format and contains annotations of: legal resources mentioned, persons, locations, organizations and time. Overlapping annotations were not allowed and only the longest named entities were annotated. The annotation of these files was enhanced with GEONAMES codes (where linking was possible). - conllup_PER_LOC_ORG_TIME Folder in which all the files are in .conllup format and contains annotations of: persons, locations, organizations and time. Overlapping annotations were not allowed and only the longest named entities were annotated. The annotation of these files was enhanced with GEONAMES codes (where linking was possible). - rdf Folder containing the corpus in RDF-Turtle format. All the annotations are available here in both span and token format. - text Folder containing the raw texts. LICENSING This work is provided under the license CC BY-NC-ND 4.0 (Attribution-NonCommercial-NoDerivatives 4.0 International). The license can be viewed online here: https://creativecommons.org/licenses/by-nc-nd/4.0/ and the full text here: https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode . CONTACT Research Institute for Artificial Intelligence "Mihai Draganescu", Romanian Academy Web: http://www.racai.ro Contact emails: vasile@racai.ro , maria@racai.ro