A simple direct-fed dual-loop antenna capable of providing hepta-band WWAN/LTE operation under surroundings an unbroken metal rim in smartphone applications is proposed. The greatest highlight this proposed that it provides a and effective multiband solution for metal-rimmed smartphone. with 5 mm height embraces the system circuit board 130 × 70 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> . Two no-ground portions 10 are set on top...

This letter aims at developing new memory architecture to overcome the daunting wall and energy issues in multicore embedded systems. We propose a heterogeneous scratch-pad (SPM) that is configured with SRAM, MRAM, Z-RAM. Based on this architecture, we two algorithms: dynamic programming (MDPDA) genetic algorithm (AGADA) allocate data different banks, therefore, reducing access cost terms of power consumption latency. Extensive experiments are performed show merits hybrid SPM effectiveness...

Ruisheng Cao, Lu Chen, Zhi Yanbin Zhao, Su Zhu, Kai Yu. Proceedings of the 59th Annual Meeting Association for Computational Linguistics and 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.

Under the condition of alternating loads, high speeds and torque levels, wear failure on gear surface is one main modes in heavy-duty-axle gears, which significantly affects its working life service performance. In this paper, shot peening method proposed to strengthen steels, effect coverage properties behavior steels systematically studied. The results show that increasing can obviously improve steels. Specifically, compared with unpeened samples, maximum compressive residual stress be...

A small-size low-profile narrow-frame antenna for seven-band WWAN/LTE operations in the internal smartphone applications is proposed. The greatest highlight of this proposed that its narrow edge has a width only 5 mm, which very novel applications. Meanwhile, small size × 40 mm <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> and low profile 3 above circuit board, makes it promising slim capable covering GSM850/900/DCS/PCS/UMTS...

Chinese short text matching usually employs word sequences rather than character to get better performance. However, segmentation can be erroneous, ambiguous or inconsistent, which consequently hurts the final To address this problem, we propose neural graph networks, a novel sentence framework capable of dealing with multi-granular input information. Instead sequence single sequence, paired lattices formed from multiple hypotheses are used as and model learns representation according an...

Zhi Chen, Lu Yanbin Zhao, Ruisheng Cao, Zihan Xu, Su Zhu, Kai Yu. Proceedings of the 2021 Conference North American Chapter Association for Computational Linguistics: Human Language Technologies. 2021.

In network link prediction, it is possible to hide a target from being predicted with small perturbation on structure. This observation may be exploited in many real world scenarios, for example, preserve privacy, or exploit financial security. There have been recent studies generate adversarial examples mislead deep learning models graph data. However, none of the previous work has considered dynamic nature real-world systems. this work, we present first study attack prediction (DNLP). The...

Unit testing is a crucial, yet often tedious and time-consuming task. To relieve developers from this burden, automated unit test generation techniques are developed. Existing tools, such as program-analysis-based tools like EvoSuite Randoop, lack program comprehension, resulting in tests with poor readability limited assertions. Language-model-based AthenaTest A3Test, have limitations the of correct tests. In paper, we introduce ChatUniTest, ChatGPT-based tool developed under...

Quality of Service (QoS) in the 6G application scenario is an important issue with premise massive data transmission. Edge caching based on fog computing network considered as a potential solution to effectively reduce content fetch delay for latency-sensitive services Internet Things (IoT) devices. Considering time-varying scenario, machine learning techniques could further by optimizing decisions. In this paper, minimize and ensure QoS network, Device-to-Device (D2D) assisted architecture...

In wire electrical discharge machining, due to the random distribution of insulating SiC particles, frequent rupture, low machining efficiency and surface quality when common brass electrode (BWE) is used process high-volume content SiCp/Al composite often appears. To address this issue, paper proposes a new preparation method zinc coating microstructure on electrodes (ZCSMWE). The ZCSMWE includes casting, coating, annealing plastic processing. experimental results show that, compared with...

Wet friction clutch is the key functional component of high-speed helicopter variable-speed transmission system, which used to change power path. In engagement process wet clutch, driving/driven disc will produce drag torque under shearing lubricating oil, reduces efficiency. This unnecessary efficiency and increases temperature. The temperature increase promotes wear gears bearings aging deformation plates, leads local service life clutch. From principle oil groove structure directly...

In any given machine learning problem, there might be many models that explain the data almost equally well. However, most algorithms return only one of these models, leaving practitioners with no practical way to explore alternative have desirable properties beyond what could expressed by a loss function. The Rashomon set is all almost-optimal models. sets can large in size and complicated structure, particularly for highly nonlinear function classes allow complex interaction terms, such as...

Abstract This paper presents an ontology-aware pretrained language model (OPAL) for end-to-end task-oriented dialogue (TOD). Unlike chit-chat models, models fulfill at least two task-specific modules: Dialogue state tracker (DST) and response generator (RG). The consists of the domain-slot-value triples, which are regarded as user’s constraints to search domain-related databases. large-scale data with annotated structured usually inaccessible. It prevents development dialogue. We propose a...