arXiv subjects

Xing Wang

Publications and source records attributed to Xing Wang.

At least 163 records · Page 9Linked to original sources

arXiv2022

MoCoViT: Mobile Convolutional Vision Transformer

Recently, Transformer networks have achieved impressive results on a variety of vision tasks. However, most of them are computationally expensive and not suitable for real-world mobile applications. In this work, we present Mobile Convolutional Vision Transformer (MoCoViT), which improves in performance and efficiency by introducing transformer into mobile convolutional networks to leverage the benefits of both architectures. Different from recent works on vision transformer, the mobile transformer block in MoCoViT is carefully designed for mobile devices and is very lightweight, accomplished through two primary modifications: the Mobile Self-Attention (MoSA) module and the Mobile Feed Forward Network (MoFFN). MoSA simplifies the calculation of the attention map through Branch Sharing scheme while MoFFN serves as a mobile version of MLP in the transformer, further reducing the computation by a large margin. Comprehensive experiments verify that our proposed MoCoViT family outperform state-of-the-art portable CNNs and transformer neural architectures on various vision tasks. On ImageNet classification, it achieves 74.5% top-1 accuracy at 147M FLOPs, gaining 1.2% over MobileNetV3 with less computations. And on the COCO object detection task, MoCoViT outperforms GhostNet by 2.1 AP in RetinaNet framework.

cs.CV↗

arXiv2022

Accurate Fruit Localisation for Robotic Harvesting using High Resolution LiDAR-Camera Fusion

Accurate depth-sensing plays a crucial role in securing a high success rate of robotic harvesting in natural orchard environments. Solid-state LiDAR (SSL), a recently introduced LiDAR technique, can perceive high-resolution geometric information of the scenes, which can be potential utilised to receive accurate depth information. Meanwhile, the fusion of the sensory information from LiDAR and camera can significantly enhance the sensing ability of the harvesting robots. This work introduces a LiDAR-camera fusion-based visual sensing and perception strategy to perform accurate fruit localisation for a harvesting robot in the apple orchards. Two SOTA extrinsic calibration methods, target-based and targetless-based, are applied and evaluated to obtain the accurate extrinsic matrix between the LiDAR and camera. With the extrinsic calibration, the point clouds and color images are fused to perform fruit localisation using a one-stage instance segmentation network. Experimental shows that LiDAR-camera achieves better quality on visual sensing in the natural environments. Meanwhile, introducing the LiDAR-camera fusion largely improves the accuracy and robustness of the fruit localisation. Specifically, the standard deviations of fruit localisation by using LiDAR-camera at 0.5 m, 1.2 m, and 1.8 m are 0.245, 0.227, and 0.275 cm respectively. These measurement error is only one one fifth of that from Realsense D455. Lastly, we have attached our visualised point cloud to demonstrate the highly accurate sensing method.

cs.RO↗

arXiv2022

Bridging the Data Gap between Training and Inference for Unsupervised Neural Machine Translation

Back-translation is a critical component of Unsupervised Neural Machine Translation (UNMT), which generates pseudo parallel data from target monolingual data. A UNMT model is trained on the pseudo parallel data with translated source, and translates natural source sentences in inference. The source discrepancy between training and inference hinders the translation performance of UNMT models. By carefully designing experiments, we identify two representative characteristics of the data gap in source: (1) style gap (i.e., translated vs. natural text style) that leads to poor generalization capability; (2) content gap that induces the model to produce hallucination content biased towards the target language. To narrow the data gap, we propose an online self-training approach, which simultaneously uses the pseudo parallel data {natural source, translated target} to mimic the inference scenario. Experimental results on several widely-used language pairs show that our approach outperforms two strong baselines (XLM and MASS) by remedying the style and content gaps.

cs.CL↗

arXiv2022

Understanding and Improving Sequence-to-Sequence Pretraining for Neural Machine Translation

In this paper, we present a substantial step in better understanding the SOTA sequence-to-sequence (Seq2Seq) pretraining for neural machine translation~(NMT). We focus on studying the impact of the jointly pretrained decoder, which is the main difference between Seq2Seq pretraining and previous encoder-based pretraining approaches for NMT. By carefully designing experiments on three language pairs, we find that Seq2Seq pretraining is a double-edged sword: On one hand, it helps NMT models to produce more diverse translations and reduce adequacy-related translation errors. On the other hand, the discrepancies between Seq2Seq pretraining and NMT finetuning limit the translation quality (i.e., domain discrepancy) and induce the over-estimation issue (i.e., objective discrepancy). Based on these observations, we further propose simple and effective strategies, named in-domain pretraining and input adaptation to remedy the domain and objective discrepancies, respectively. Experimental results on several language pairs show that our approach can consistently improve both translation performance and model robustness upon Seq2Seq pretraining.

cs.CL↗

arXiv2022

Factorization and Sudakov Resummation in Leptonic Radiative $B$ Decay -- A Reappraisal

The $B$-meson light-cone distribution amplitude is an important non-perturbative quantity arising in the factorization of the amplitudes for many exclusive decays of $B$ mesons, such as $B^-\toγ\,\ell^-\barν$. We reconsider the renormalization-group (RG) equation satisfied by this function and present its solution at next-to-leading order (NLO) in RG-improved perturbation theory in Laplace space and, for the first time, in momentum space and the so-called diagonal (or dual) space. Since the information needed to describe the $B$ decay processes at leading order in $Λ_{\rm QCD}/m_b$ is most directly contained in the distribution amplitude in Laplace space evaluated near the origin, we propose an unbiased parameterization of this object in terms of a small set of uncorrelated hadronic parameters. Using recent results on the three-loop anomalous dimension for heavy-light current operators, we derive an expression for the convolution integral appearing in the $B^-\toγ\,\ell^-\barν$ factorization formula that is explicitly scale independent, and we evaluate this formula at (approximate) NNLO.

hep-ph↗

arXiv2022

WIMP Dark Matter at High Energy Muon Colliders $-$A White Paper for Snowmass 2021

In a previous publication, we showed that a high energy muon collider can make decisive statements about the electroweak (WIMP) Dark Matter (DM), reaching a DM mass which could give the observed thermal relic abundance. In this document, we report new studies of the spin-$0$ minimal WIMP DM at high energy muon colliders, and update our results on the fermionic spin-$1/2$ case. We find that, by combining multiple inclusive missing mass search channels, it is possible to fully cover the thermal targets of fermionic and scalar doublets, and Dirac triplet, with a 10 TeV muon collider. Higher energies, 14 TeV$-$30 TeV, would be able to cover the thermal targets of Majorana and scalar triplet. For direct discovery of the higher EW multiplets with $n\geq5$, one may need to go beyond a 30 TeV muon collider to fully cover their thermal mass expectation.

hep-ph↗

arXiv2022

Axion-Like Particles at High Energy Muon Colliders -- A White paper for Snowmass 2021

We study the discovery potential for heavy axion-like particles (ALPs) and the perspectives for determining their coupling properties at a muon collider. Focusing on their couplings to the Standard Model (SM) gauge bosons $γ, Z, W^\pm$, we show that a high-energy muon collider can substantially extend the mass coverage, essentially reaching the kinematic limit of the collider energy. The unique kinematics allow for non-ambiguous determination of the individual coupling strengths. The associated production via $μ^+μ^-$ annihilation and the VBF processes with the tagged outgoing muons can be utilized to verify the CP property of the ALPs. We illustrate our results for a muon collider running at 3 TeV and 10 TeV.

hep-ph↗

arXiv2022

Topological semimetal phases in a family of monolayer X3YZ6 (X=Nb,Ta, Y=Si,Ge,Sn, Z=S,Se,Te) with abundant nodal lines and nodes

The electronic and topological properties of single-layer X3YZ6 (X=Nb,Ta, Y=Si,Ge,Sn, Z=S,Se,Te) materials have been studied with the aid of first principles calculations. This kind of materials belong to topological semimetals (TMs) with abundant nodal lines and nodes. Considering their similar properties, we focus on the analysis of Ta3SnTe6 and Ta3SiSe6. The present of spin-orbit coupling (SOC) leads to the transition from type-I nodal lines to Dirac points as well as the disappear of type-II Dirac points. The three-dimensional (3D) band diagrams reproduce vividly the characteristics of nodes and nodal lines. The appearance of the flat bands in (110) edge states further confirm their nontrivial topological properties. We also explore the relationship among different nodal lines (nodes), crystal symmetry and SOC. The type-I nodal lines are protected by Mz and My symmetry in the absent of SOC. Symmetry breaking leads to band splitting even in the presence of SOC. The single-layer X3YZ6 can be used as candidates for two-dimensional (2D) TMs and provide a platform for further study of interesting physical phenomena.

cond-mat.mtrl-sci↗

arXiv2022

Proximity-Induced Superconductivity in Atomically Precise Nanographene

Obtaining a robust superconducting state in atomically precise nanographene (NG) structures by proximity to a superconductor could foster the discovery of topological superconductivity in graphene. On-surface synthesis of such NGs has been achieved on noble metals or metal oxides, however, it is still absent on superconductors. Here, we present a synthetic method to induce superconductivity to polymeric chains and NGs adsorbed on the superconducting Nb(110) substrate covered by thin Ag films. Using atomic force microscopy at low-temperature, we characterize the chemical structure of each sub-product formed on the superconducting Ag layer. Scanning tunneling spectroscopy further allows us to elucidate electronic properties of these nanostructures, which consistently show a superconducting gap. We foresee our approach to become a promising platform for exploring the interplay between carbon magnetism and superconductivity at the fundamental level.

cond-mat.supr-con↗

arXiv2021

Soft Robotic Finger with Variable Effective Length enabled by an Antagonistic Constraint Mechanism

Compared to traditional rigid robotics, soft robotics has attracted increasing attention due to its advantages as compliance, safety, and low cost. As an essential part of soft robotics, the soft robotic gripper also shows its superior while grasping the objects with irregular shapes. Recent research has been conducted to improve its grasping performance by adjusting the variable effective length (VEL). However, the VEL achieved by multi-chamber design or tunable stiffness shape memory material requires complex pneumatic circuit design or a time-consuming phase-changing process. This work proposes a fold-based soft robotic actuator made from 3D printed filament, NinjaFlex. It is experimentally tested and represented by the hyperelastic model. Mathematic and finite element modelling is conducted to study the bending behaviour of the proposed soft actuator. Besides, an antagonistic constraint mechanism is proposed to achieve the VEL, and the experiments demonstrate that better conformity is achieved. Finally, a two-mode gripper is designed and evaluated to demonstrate the advances of VEL on grasping performance.

cs.RO↗

arXiv2021

Progressive Automatic Design of Search Space for One-Shot Neural Architecture Search

Neural Architecture Search (NAS) has attracted growing interest. To reduce the search cost, recent work has explored weight sharing across models and made major progress in One-Shot NAS. However, it has been observed that a model with higher one-shot model accuracy does not necessarily perform better when stand-alone trained. To address this issue, in this paper, we propose Progressive Automatic Design of search space, named PAD-NAS. Unlike previous approaches where the same operation search space is shared by all the layers in the supernet, we formulate a progressive search strategy based on operation pruning and build a layer-wise operation search space. In this way, PAD-NAS can automatically design the operations for each layer and achieve a trade-off between search space quality and model diversity. During the search, we also take the hardware platform constraints into consideration for efficient neural network model deployment. Extensive experiments on ImageNet show that our method can achieve state-of-the-art performance.

cs.CV↗

arXiv2021

Geometry-Aware Fruit Grasping Estimation for Robotic Harvesting in Orchards

Field robotic harvesting is a promising technique in recent development of agricultural industry. It is vital for robots to recognise and localise fruits before the harvesting in natural orchards. However, the workspace of harvesting robots in orchards is complex: many fruits are occluded by branches and leaves. It is important to estimate a proper grasping pose for each fruit before performing the manipulation. In this study, a geometry-aware network, A3N, is proposed to perform end-to-end instance segmentation and grasping estimation using both color and geometry sensory data from a RGB-D camera. Besides, workspace geometry modelling is applied to assist the robotic manipulation. Moreover, we implement a global-to-local scanning strategy, which enables robots to accurately recognise and retrieve fruits in field environments with two consumer-level RGB-D cameras. We also evaluate the accuracy and robustness of proposed network comprehensively in experiments. The experimental results show that A3N achieves 0.873 on instance segmentation accuracy, with an average computation time of 35 ms. The average accuracy of grasping estimation is 0.61 cm and 4.8$^{\circ}$ in centre and orientation, respectively. Overall, the robotic system that utilizes the global-to-local scanning and A3N, achieves success rate of harvesting ranging from 70\% - 85\% in field harvesting experiments.

cs.RO↗

arXiv2021

Ultrafast formation of transient 2D diamond-like structure in twisted bilayer graphene

Due to the absence of matching carbon atoms at honeycomb centers with carbon atoms in adjacent graphene sheets, theorists predicted that a sliding process is needed to form AA, AB, or ABC stacking when directly converting graphite into sp3 bonded diamond. Here, using twisted bilayer graphene, which naturally provides AA and AB stacking configurations, we report the ultrafast formation of a transient 2D diamond-like structure (which is not observed in aligned graphene) under femtosecond laser irradiation. This photo-induced phase transition is evidenced by the appearance of new bond lengths of 1.94A and 3.14A in the time-dependent differential pair distribution function using MeV ultrafast electron diffraction. Molecular dynamics and first principles calculation indicate that sp3 bonds nucleate at AA and AB stacked areas in moire pattern. This work sheds light on the direct graphite-to-diamond transformation mechanism, which has not been fully understood for more than 60 years.

cond-mat.mtrl-sci↗

arXiv2021

Radiative Quark Jet Function with an External Gluon

Factorization theorems in soft-collinear effective theory at subleading order in power counting involve ``radiative jet functions'', defined in terms of matrix elements of collinear fields with a soft momentum emitted from inside the jet. Of particular importance are the radiative quark jet functions with an external photon or gluon, which arise e.g. in the factorization theorems for the Higgs-boson amplitudes $h\toγγ$, $h\to gg$ and $gg\to h$ induced by light-quark loops. While the photon case has been studied extensively in previous work, we present here a detailed study of the radiative jet function with an external gluon. We calculate this jet function at one- and two-loop order, derive its one-loop anomalous dimension and study its renormalization-group evolution.

hep-ph↗

arXiv2021

Adaptive Multi-receptive Field Spatial-Temporal Graph Convolutional Network for Traffic Forecasting

Mobile network traffic forecasting is one of the key functions in daily network operation. A commercial mobile network is large, heterogeneous, complex and dynamic. These intrinsic features make mobile network traffic forecasting far from being solved even with recent advanced algorithms such as graph convolutional network-based prediction approaches and various attention mechanisms, which have been proved successful in vehicle traffic forecasting. In this paper, we cast the problem as a spatial-temporal sequence prediction task. We propose a novel deep learning network architecture, Adaptive Multi-receptive Field Spatial-Temporal Graph Convolutional Networks (AMF-STGCN), to model the traffic dynamics of mobile base stations. AMF-STGCN extends GCN by (1) jointly modeling the complex spatial-temporal dependencies in mobile networks, (2) applying attention mechanisms to capture various Receptive Fields of heterogeneous base stations, and (3) introducing an extra decoder based on a fully connected deep network to conquer the error propagation challenge with multi-step forecasting. Experiments on four real-world datasets from two different domains consistently show AMF-STGCN outperforms the state-of-the-art methods.

cs.LG↗

arXiv2021

Failure-averse Active Learning for Physics-constrained Systems

Active learning is a subfield of machine learning that is devised for design and modeling of systems with highly expensive sampling costs. Industrial and engineering systems are generally subject to physics constraints that may induce fatal failures when they are violated, while such constraints are frequently underestimated in active learning. In this paper, we develop a novel active learning method that avoids failures considering implicit physics constraints that govern the system. The proposed approach is driven by two tasks: the safe variance reduction explores the safe region to reduce the variance of the target model, and the safe region expansion aims to extend the explorable region exploiting the probabilistic model of constraints. The global acquisition function is devised to judiciously optimize acquisition functions of two tasks, and its theoretical properties are provided. The proposed method is applied to the composite fuselage assembly process with consideration of material failure using the Tsai-wu criterion, and it is able to achieve zero-failure without the knowledge of explicit failure regions.

stat.ML↗

arXiv2021

A Tactile-enabled Grasping Method for Robotic Fruit Harvesting

In the robotic crop harvesting environment, foreign objects intrusion in the gripper workspace is frequently occurring and unignorable, however, rarely addressed. This paper presents a novel intelligent robotic grasping method capable of handling obstacle interference, which is the first of its kind in the literature. The proposed method combines deep learning algorithms with low-cost tactile sensing hardware on a multi-DoF soft robotic gripper. Through experimental validations, the proposed method demonstrated promising performance in distinguishing various grasping scenarios. The 4-finger independently controlled gripper presented outstanding adaptability to handle various picking scenarios. The overall performance of this work indicated great potential for solving the robotic fruit harvesting challenges.

cs.RO↗

arXiv2021

Next-to-leading power SCET in Higgs amplitudes induced by light quarks

In this article, we report how to use the "plus-type" subtraction scheme to deal with endpoint divergences in $h\rightarrowγγ$ or $gg\rightarrow h$ amplitudes induced by light quark loop, which can be formulated by next-to-leading power (NLP) SCET. This subtraction is ensured by two re-factorization conditions, which have been proven to all orders in $α_s$. Based on these conditions, cutoffs emerge naturally after some re-arrangements to handle endpoint divergences and renormalization is compatible with that. Our formalism can analytically reproduce three-loop amplitudes up to $\ln^3(-M_h^2/m_b^2-i0)$ and resum to next-to-leading logarithm and beyond.

hep-ph↗