No abstract

Here, we design and engineer an axially asymmetric GaAs/AlGaAs/GaAs (G/A/G) nanowire (NW) photodetector that operates efficiently at room temperature. Based on the I-type band structure, the device can realize a two-dimensional electron–hole tube (2DEHT) structure for the substantial performance enhancement. The 2DEHT is observed to form at the interface on both sides of GaAs/AlGaAs barriers, which constructs effective pathways for both electron and hole transport in reducing the photocarrier recombination and enhancing the device photocurrent. In particular, the G/A/G NW photodetector exhibits a responsivity of 0.57 A/W and a detectivity of 1.83 × 1010 Jones, which are about 7 times higher than those of the pure GaAs NW device. The recombination probability has also been significantly suppressed from 81.8% to 13.2% with the utilization of the 2DEHT structure. All of these can evidently demonstrate the importance of the appropriate band structure design to promote photocarrier generation, separation, and collection for high-performance optoelectronic devices.

We study the problem of attacking video recognition models in the black-box setting, where the model information is unknown and the adversary can only make queries to detect the predicted top-1 class and its probability. Compared with the black-box attack on images, attacking videos is more challenging as the computation cost for searching the adversarial perturbations on a video is much higher due to its high dimensionality. To overcome this challenge, we propose a heuristic black-box attack model that generates adversarial perturbations only on the selected frames and regions. More specifically, a heuristic-based algorithm is proposed to measure the importance of each frame in the video towards generating the adversarial examples. Based on the frames' importance, the proposed algorithm heuristically searches a subset of frames where the generated adversarial example has strong adversarial attack ability while keeps the perturbations lower than the given bound. Besides, to further boost the attack efficiency, we propose to generate the perturbations only on the salient regions of the selected frames. In this way, the generated perturbations are sparse in both temporal and spatial domains. Experimental results of attacking two mainstream video recognition methods on the UCF-101 dataset and the HMDB-51 dataset demonstrate that the proposed heuristic black-box adversarial attack method can significantly reduce the computation cost and lead to more than 28% reduction in query numbers for the untargeted attack on both datasets.

Nanowire photodetectors, which have the advantages of fast response and high photoelectric conversion efficiency, can be widely applied in various industries. However, the rich surface states result in large dark current and can hinder the development of high-performance nanowire photodetectors. In this paper, the influence and mechanism of sulfur surface passivation on the dark current of a single GaAs nanowire photodetector have been studied. The dark current is significantly reduced by about 30 times after surface passivation. We confirm that the origin of the reduction of dark current is the decrease in the surface state density. As a result, a single GaAs nanowire photodetector with low dark current of 7.18 × 10 pA and high detectivity of 9.04 × 10 cmHzW has been achieved. A simple and convenient way to realize high-performance GaAs-based photodetectors has been proposed.