Con-Detect: Detecting adversarially perturbed natural language inputs to deep classifiers through holistic analysis by Hassan, Ali

“…Even in the worst case, we note a 100% increase in the required number of queries and a 50% increase in the number of words perturbed, suggesting that Con-Detect is hard to evade.…”
Get full text
Get full text
Get full text

Proteome analysis of Candida albicans cell wall mutants. (c2011) by Koussa, Joseph Wissam

Get full text
Get full text

Healthcare Services for Spectators during the FIFA World Cup 2022^TM by Muhammad Arsalan Zamir (18466355)

“…Qatar received 1.4 million visitors during the FIFA World Cup 2022^TM, a 50% increase in population. The aim was to provide high quality medical care to the visiting spectators, while continuing regular medical services for residents. …”

Con-Detect: Detecting Adversarially Perturbed Natural Language Inputs to Deep Classifiers Through Holistic Analysis by Hassan Ali (3348749)

“…Even in the worst case, we note a 100% increase in the required number of queries and a 50% increase in the number of words perturbed, suggesting that Con-Detect is hard to evade. …”

Management of leak after sleeve gastrectomy: outcomes of 73 cases, treatment algorithm and predictors of resolution by Moataz Bashah (14148867)

“…Patients on jejunal feeding had shorter resolution time (HR = 2.7, P = 0.018), while patients on total parenteral nutrition and post-endoscopic dilatation had 66% and 50% increases in the resolution time; (HR = 0.34, P = 0.026) and (HR = 0.5, <i>P</i> = 0.047), respectively.…”