As virtual home assistants are becoming more popular, there is an emerging need for supporting languages other than English. While more wide-spread or popular languages such as Spanish, French or Hindi are already integrated into existing home assistants like Google Home or Alexa, integration of other less-known languages such as Romanian is still missing. This paper explores the problem of Natural Language Understanding (NLU) applied to a Romanian home assistant. We propose a customized capsule neural network architecture that performs intent detection and slot filling in a joint manner and we evaluate how well it handles utterances containing various levels of complexity. The capsule network model shows a significant improvement in intent detection when compared to models built using the well-known Rasa NLU tool. Through error analysis, we observe clear error patterns that occur systematically. Variability in language when expressing one intent proves to be the biggest challenge encountered by the model.
This paper explores the use of control-flow information for detecting IoT malware across three different platforms. We propose a pipeline which extracts Control-Flow Graphs (CFGs) from program binaries, projects them onto an embedding space, then computes the similarity between the programs, in an attempt to explore whether this approach could help identify malware families across different platforms. We explore three different methods for generating graph embeddings – at node, edge and graph level – and find the latter to produce the best discrimination between malware families across different platforms. Moreover, we propose a computationally efficient method for computing graph similarities for graphs with multiple connected components, which achieves accuracy on par with the cubic Hungarian Algorithm matching approach. The evaluations performed on a dataset comprising of clean and malware samples belonging to 18 different families yielded that our CFG embedding similarity approach can easily identify most malware family samples belonging to the same family, for the same platform, obtaining AUC scores above 90%. The method shows strong potential for discriminating between malware families across different platforms. We are currently exploring cross-platform limitations and improvement points, and we share preliminary insights from our current work towards the end of the paper.
Dot-product attention is a powerful mechanism for capturing contextual information. Models that build on top of it have acclaimed state-of-the-art performance in various domains, ranging from sequence modelling to visual tasks. However, the main bottleneck is the construction of the attention map, which is quadratic with respect to the number of tokens in the sequence. Consequently, efficient alternatives have been developed in parallel, but it was only recently that their performances were compared and contrasted. This study performs a comparative analysis between some efficient attention mechanisms in the context of a purely attention-based spatio-temporal forecasting model used for traffic prediction. Experiments show that these methods can reduce the training times by up to 28% and the inference times by up to 31%, while the performance remains on par with the baseline.