TweepFake: About detecting deepfake tweets.

The recent advances in language modeling significantly improved the generative capabilities of deep neural models: in 2019 OpenAI released GPT-2, a pre-trained language model that can autonomously generate coherent, non-trivial and human-like text samples. Since then, ever more powerful text generative models have been developed. Adversaries can exploit these tremendous generative capabilities to enhance social bots that will have the ability to write plausible deepfake messages, hoping to contaminate public debate.

Does monitoring of pre-/post-dialyzer pressure difference improve efficiency in intermittent hemodialysis?

Background: Continuous monitoring of pre-/post-dialyzer pressure difference (DeltaP) is widely used in continuous renal replacement therapies to monitor extracorporeal circuit function. The aim of this study was to verify whether DeltaP may help to identify chronic subclinical worsening of dialysis quality due to incomplete dialyzer clotting in intermittent hemodialysis.

Methods: Nine chronic hemodialysis patients were enrolled in the study and dialyzed twice (high-flux polysulfone dialyzer) with DeltaP and urea-clearance monitoring: the first session with a standard anticoagulation and the second without.