Ključne besede: natural language processing, language models, sarcasm detection, transformer architectureCelotno besedilo (datoteka, 286,70 KB)
Ključne besede: question answering, cross-lingual transfer, information retrieval, deep learning, Macedonian language, transformer modelsCelotno besedilo (datoteka, 959,65 KB)
Ključne besede: text anonymization, Slovenian, de-identification, deep learning, transformer modelsCelotno besedilo (datoteka, 719,07 KB)
Ključne besede: natural language processing, machine learning, transformer neural networks, Bayesian neural networks, BERT models, prediction uncertainty, reliability estimation, Monte Carlo dropout, Bayesian BERT, sentic computing, model calibrationCelotno besedilo (datoteka, 3,06 MB) Gradivo ima več datotek! Več...