The Effectiveness of Dynamically Processed Incremental Descriptions in Human Robot Interaction

We explore the effectiveness of a dynamically processed incremental referring description system using under-specified ambiguous descriptions that are then built upon using linguistic repair statements, which we refer to as a dynamic system. We build a dynamically processed incremental referring description generation system that is able to provide contextual navigational statements to describe an object in a potential real-world situation of nuclear waste sorting and maintenance. In a study of 31 participants, we test the dynamic system in a case where a user is remote operating a robot to sort nuclear waste, with the robot assisting them in identifying the correct barrels to be removed. We compare these against a static non-ambiguous description given in the same scenario. As well as looking at efficiency with time and distance measurements, we also look at user preference. Results show that our dynamic system was a much more efficient method—taking only 62% of the time on average—for finding the correct barrel. Participants also favoured our dynamic system.

Jittering: A computationally efficient method for generating realistic route networks from origin-destination data

Origin-destination (OD) data is a vital source of information on travel patterns but its utility is limited by reliance on zone centroids. This paper presents a reproducible and open two-stage ‘jittering’ approach to tackling this problem, which (1) uses random points to represent unique start and end points (sampling), and (2) splits OD pairs representing many trips into many ‘sub-OD’ pairs. We find that route networks generated from jittered OD data are more diffuse and potentially realistic based on an example from Edinburgh. Further work is needed to validate the approach and to find optimal parameters for sampling and disaggregation.

Effect of Microwave-Assisted Aqueous Two-Phase Extraction of α-Solanine from S. retroflexum and Analysis on UHPLC-qTOF-MS

A new, fast and efficient method, hyphenated microwave-assisted aqueous two-phase extraction (MA-ATPE) was applied in the extraction of α-solanine from Solanum retroflexum. This environmentally friendly extraction method applied water and ethanol as extraction solvents. Central composite design (CCD) was performed which included numerical parameters such as time, mass of plant powder and microwave power. The categorical factors included the chaotrope — NaCl or the kosmotrope — Na2CO3. Fitting the central composite design response surface model to the data generated a quadratic model with a good fit (R2 = 0.920). The statistically significant (p < 0.05) parameters such as time and mass of plant powder were influential in the extraction of α-solanine. Quantification of α-solanine was achieved using a robust and sensitive feature of the ultra-high performance quadrupole time of flight mass spectrometer (UHPLC-qTOF-MS), multiple reaction monitoring (MRM). The optimized condition for the extraction of α-solanine in the presence of NaCl and Na2CO3 was a period of 1 min at a mass of 1.2 g using a microwave power of 40%. Maximal extraction of α-solanine was 93.50 mg kg−1 and 72.16 mg kg−1 for Na2CO3 and NaCl, respectively. The synergistic effect of salting-out and microwave extraction was influential in extraction of α-solanine. Furthermore, the higher negative charge density of the kosmotrope (Na2CO3) was responsible for its greater extraction of α-solanine than chaotrope (NaCl). The shorter optimal extraction times of MA-ATPE make it a potential technique that could meet market demand as it is a quick, green and efficient method for removal of toxic metabolites in nutraceuticals.

UV-Light-Induced Dehydrogenative N-Acylation of Amines with 2-Nitrobenzaldehydes to Give 2-Aminobenzamides

A simple, mild, green and efficient method for the synthesis of 2-aminobenzamides was highly desired in organic synthesis. Herein, we developed an efficient, one-pot strategy for the synthesis of 2-aminobenzamides with high yields irradiated by UV light. 32 examples proceeded successfully by this photo-induced protocol. The yield reached up to 92%. The gram scale was also achieved easily. This building block could be applied in the preparation of quinazolinones derivatives. Amino acid derivatives could be employed smoothly at room temperature. Finally, a plausible mechanism was proposed.

Knockout of circRNAs by base editing back-splice sites of circularized exons

Many circular RNAs (circRNAs) are produced from back-splicing of exons of precursor mRNAs and are generally co-expressed with cognate linear RNAs. Methods for circRNA-specific knockout are lacking, largely due to sequence overlaps between forms. Here, we use base editors (BEs) for circRNA depletion. By targeting splice sites involved in both back-splicing and canonical splicing, BEs can repress circular and linear RNAs. Targeting sites predominantly for circRNA biogenesis, BEs could efficiently repress the production of circular but not linear RNAs. As hundreds of exons are predominantly back-spliced to produce circRNAs, this provides an efficient method to deplete circRNAs for functional study.

Highly Efficient and Ambient-Temperature Synthesis of Benzimidazoles via Co(III)/Co(II)-Mediated Redox Catalysis

An efficient method for ambient-temperature synthesis of a variety of 2-substituted and 1,2-disubstituted benzimidazoles from aldehyde and phenylenediamine substrates has been developed by utilizing Co(III)/Co(II)-mediated redox catalysis. The combination of only 1 mol% of Co(acac)2 and stoichiometric amount of hydrogen peroxide provides a fast, green, and mild access to a diversity of benzimidazoles under solvent-free conditions.

An Efficient Method for Generating Adversarial Malware Samples

Deep learning methods have been applied to malware detection. However, deep learning algorithms are not safe, which can easily be fooled by adversarial samples. In this paper, we study how to generate malware adversarial samples using deep learning models. Gradient-based methods are usually used to generate adversarial samples. These methods generate adversarial samples case-by-case, which is very time-consuming to generate a large number of adversarial samples. To address this issue, we propose a novel method to generate adversarial malware samples. Different from gradient-based methods, we extract feature byte sequences from benign samples. Feature byte sequences represent the characteristics of benign samples and can affect classification decision. We directly inject feature byte sequences into malware samples to generate adversarial samples. Feature byte sequences can be shared to produce different adversarial samples, which can efficiently generate a large number of adversarial samples. We compare the proposed method with the randomly injecting and gradient-based methods. The experimental results show that the adversarial samples generated using our proposed method have a high successful rate.