Gervásio PT-* is a foundation, large language model for the Portuguese language. It is a decoder of the GPT family, based on the neural architecture Transformer and developed over the Pythia model, with competitive performance for this language. It has different versions that were trained for ...
The DEEB Corpus contains the transcriptions of 1200 narrative texts written by pupils in their 4th, 6th and 9th year Portuguese Language exams in the public school system in Portugal.
The Portuguese Parliamentary Corpus is part of the Mutlilingual ParlaMint Corpus, a set of comparable corpora containing transcriptions of parliamentary debates of 29 European countries and autonomous regions. The Portuguese corpus (ParlaMint-PT) comprehends transcripts of sessions in the time pe...
Grafone-Tool is a tool for conversion from grapheme to phoneme for European Portuguese. The converter works with the Portuguese spelling, both prior to and after the Orthographic Agreement of 1990.
ViPER is a verb lexical database with +7,000 verb senses, along with their structural, distributional, and transformational properties. The verb senses are classified based on the main syntactic properties of their construction. Around 70 formal classes have been devised. For each verb sense, its...
Dicionário de Gentílicos e Topónimos is a list of pairs of toponyms and demonyms. The toponyms and demonyms included have a morphologically compositional relation between each other. The list contains around 1500 such pairs and additionally provides information on the toponym referent (upper unit...
A Portuguese as a non-native language learners' corpus of written texts with three independent subcorpora: - Portuguese as a Foreign Language: Subcorpus Português Língua Estrangeira (PEAPL2_PLE) http://teitok2.iltec.pt/peapl2-ple/index.php?action=home - East Timorese Portuguese: Subcorpus T...
The LT Corpus (Literary Corpus) contains approximately 1,781,083 running words of European and Brazilian Portuguese. It includes 70 copyright-free classics (61 Portugal and 9 from Brazil) published before 1940.
LX-AP was created from the translation of Almuhareb-Poesio (ap) benchmark (Almuhareb and Poesio, 2005). The original data set was created considering three aspects: POS, frequency and ambiguity. It contains 402 names from 21 categories of WordNet, with 13 to 21 names from each one of those categ...
The test set described in was used as the basis for the assessment of word embeddings. An example entry in this data set would read: ‘Berlin Germany Lisbon Portugal’. With these four words relations – as in this example – one can test semantic analogies by using any of the possible combinations o...