Health-Related Content in Transformer-Based Language Models: Exploring Bias in Domain General vs. Domain Specific Training Sets.

In this communication, we demonstrate that the bias observed in domain general training sets with health-related content is not improved in domain specific health-communication corpora, contra.

Health-Related Content in Transformer-Based Deep Neural Network Language Models: Exploring Cross-Linguistic Syntactic Bias.

This paper explores a methodology for bias quantification in transformer-based deep neural network language models for Chinese, English, and French. When queried with health-related mythbusters on COVID-19, we observe a bias that is not of a semantic/encyclopaedical knowledge nature, but rather a syntactic one, as predicted by theoretical insights of structural complexity. Our results highlight the need for the creation of health-communication corpora as training sets for deep learning.

New trends in the cyber and street market of recreational drugs? The case of 2C-T-7 ('Blue Mystic').

2C-T-7 ('Blue Mystic'), an illicit compound which shows similarities with MDMA and other designer drugs, has been only occasionally identified in the EU, but discussion on the Internet between experimenters has recently grown significantly. We aimed at collecting together in a review the available information on 2C-T-7, both at the cyber and at the street market level. 2C-T-7 was first synthesized in 1986; its desired effects include both a sense of empathy and of well-being.