Zobrazeno 1 - 9
of 9
pro vyhledávání: '"Pires, Telmo"'
Autor:
Colombo, Pierre, Pires, Telmo, Boudiaf, Malik, Melo, Rui, Culver, Dominic, Morgado, Sofia, Malaboeuf, Etienne, Hautreux, Gabriel, Charpentier, Johanne, Desa, Michael
In this paper, we introduce SaulLM-54B and SaulLM-141B, two large language models (LLMs) tailored for the legal sector. These models, which feature architectures of 54 billion and 141 billion parameters, respectively, are based on the Mixtral archite
Externí odkaz:
http://arxiv.org/abs/2407.19584
Autor:
Colombo, Pierre, Pires, Telmo Pessoa, Boudiaf, Malik, Culver, Dominic, Melo, Rui, Corro, Caio, Martins, Andre F. T., Esposito, Fabrizio, Raposo, Vera Lúcia, Morgado, Sofia, Desa, Michael
In this paper, we introduce SaulLM-7B, a large language model (LLM) tailored for the legal domain. With 7 billion parameters, SaulLM-7B is the first LLM designed explicitly for legal text comprehension and generation. Leveraging the Mistral 7B archit
Externí odkaz:
http://arxiv.org/abs/2403.03883
The Transformer architecture has two main non-embedding components: Attention and the Feed Forward Network (FFN). Attention captures interdependencies between words regardless of their position, while the FFN non-linearly transforms each input token
Externí odkaz:
http://arxiv.org/abs/2309.01826
Multilingual Machine Translation promises to improve translation quality between non-English languages. This is advantageous for several reasons, namely lower latency (no need to translate twice), and reduced error cascades (e.g., avoiding losing gen
Externí odkaz:
http://arxiv.org/abs/2305.02665
Structured State Spaces for Sequences (S4) is a recently proposed sequence model with successful applications in various tasks, e.g. vision, language modeling, and audio. Thanks to its mathematical formulation, it compresses its input to a single hid
Externí odkaz:
http://arxiv.org/abs/2304.12776
Non-autoregressive approaches aim to improve the inference speed of translation models by only requiring a single forward pass to generate the output sequence instead of iteratively producing each predicted token. Consequently, their translation qual
Externí odkaz:
http://arxiv.org/abs/2205.10577
Autor:
Weller, Orion, Sperber, Matthias, Pires, Telmo, Setiawan, Hendra, Gollan, Christian, Telaar, Dominic, Paulik, Matthias
Code switching (CS) refers to the phenomenon of interchangeably using words and phrases from different languages. CS can pose significant accuracy challenges to NLP, due to the often monolingual nature of the underlying systems. In this work, we focu
Externí odkaz:
http://arxiv.org/abs/2204.05076
In this paper, we show that Multilingual BERT (M-BERT), released by Devlin et al. (2018) as a single language model pre-trained from monolingual corpora in 104 languages, is surprisingly good at zero-shot cross-lingual model transfer, in which task-s
Externí odkaz:
http://arxiv.org/abs/1906.01502
Dissertação de mestrado integrado em Engenharia Mecânica
Hoje em dia há uma crescente preocupação por parte dos consumidores em eliminar, reduzir ou conter custos associados aos elevados consumos de energia e à proteção do meio ambiente
Hoje em dia há uma crescente preocupação por parte dos consumidores em eliminar, reduzir ou conter custos associados aos elevados consumos de energia e à proteção do meio ambiente
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=od_______307::d96208c300c9fc6b04136ce6535df8df
https://hdl.handle.net/1822/77271
https://hdl.handle.net/1822/77271