arXiv subjects

Jing Lu

Publications and source records attributed to Jing Lu.

At least 127 records · Page 7Linked to original sources

arXiv2022

On the approximation of queue-length distributions in transportation networks

This paper focuses on the analytical probabilistic modeling of vehicular traffic. It formulates a stochastic node model. It then formulates a network model by coupling the node model with the link model of Lu and Osorio (2018), which is a stochastic formulation of the traffic-theoretic link transmission model. The proposed network model is scalable and computationally efficient, making it suitable for urban network optimization. For a network with $r$ links, each of space capacity $\ell$, the model has a complexity of $\mathcal{O}(r\ell)$. The network model yields the marginal distribution of link states. The model is validated versus a simulation-based network implementation of the stochastic link transmission model. The validation experiments consider a set of small network with intricate traffic dynamics. For all scenarios, the proposed model accurately captures the traffic dynamics. The network model is used to address a signal control problem. Compared to the probabilistic link model of Lu and Osorio (2018) with an exogenous node model and a benchmark deterministic network loading model, the proposed network model derives signal plans with better performance. The case study highlights the added value of using between-link (i.e., across-node) interaction information for traffic management and accounting for stochasticity in the network.

cs.GT↗

arXiv2022

From Data to Software to Science with the Rubin Observatory LSST

The Vera C. Rubin Observatory Legacy Survey of Space and Time (LSST) dataset will dramatically alter our understanding of the Universe, from the origins of the Solar System to the nature of dark matter and dark energy. Much of this research will depend on the existence of robust, tested, and scalable algorithms, software, and services. Identifying and developing such tools ahead of time has the potential to significantly accelerate the delivery of early science from LSST. Developing these collaboratively, and making them broadly available, can enable more inclusive and equitable collaboration on LSST science. To facilitate such opportunities, a community workshop entitled "From Data to Software to Science with the Rubin Observatory LSST" was organized by the LSST Interdisciplinary Network for Collaboration and Computing (LINCC) and partners, and held at the Flatiron Institute in New York, March 28-30th 2022. The workshop included over 50 in-person attendees invited from over 300 applications. It identified seven key software areas of need: (i) scalable cross-matching and distributed joining of catalogs, (ii) robust photometric redshift determination, (iii) software for determination of selection functions, (iv) frameworks for scalable time-series analyses, (v) services for image access and reprocessing at scale, (vi) object image access (cutouts) and analysis at scale, and (vii) scalable job execution systems. This white paper summarizes the discussions of this workshop. It considers the motivating science use cases, identified cross-cutting algorithms, software, and services, their high-level technical specifications, and the principles of inclusive collaborations needed to develop them. We provide it as a useful roadmap of needs, as well as to spur action and collaboration between groups and individuals looking to develop reusable software for early LSST science.

astro-ph.IM↗

arXiv2022

Rapid-Flooding Time Synchronization for Large-Scale Wireless Sensor Networks

Accurate and fast-convergent time synchronization is very important for wireless sensor networks. The flooding time synchronization converges fast, but its transmission delay and by-hop error accumulation seriously reduce the synchronization accuracy. In this article, a rapidflooding multiple one-way broadcast time-synchronization (RMTS) protocol for large-scale wireless sensor networks is proposed. To minimize the by-hop error accumulation, the RMTS uses maximum likelihood estimations for clock skew estimation and clock offset estimation, and quickly shares the estimations among the networks. As a result, the synchronization error resulting from delays is greatly reduced, while faster convergence and higher-accuracy synchronization is achieved. Extensive experimental results demonstrate that, even over 24-hops networks, the RMTS is able to build accurate synchronization at the third synchronization period, and moreover, the by-hop error accumulation is slower when the network diameter increases.

eess.SY↗

arXiv2022

Inference skipping for more efficient real-time speech enhancement with parallel RNNs

Deep neural network (DNN) based speech enhancement models have attracted extensive attention due to their promising performance. However, it is difficult to deploy a powerful DNN in real-time applications because of its high computational cost. Typical compression methods such as pruning and quantization do not make good use of the data characteristics. In this paper, we introduce the Skip-RNN strategy into speech enhancement models with parallel RNNs. The states of the RNNs update intermittently without interrupting the update of the output mask, which leads to significant reduction of computational load without evident audio artifacts. To better leverage the difference between the voice and the noise, we further regularize the skipping strategy with voice activity detection (VAD) guidance, saving more computational load. Experiments on a high-performance speech enhancement model, dual-path convolutional recurrent network (DPCRN), show the superiority of our strategy over strategies like network pruning or directly training a smaller model. We also validate the generalization of the proposed strategy on two other competitive speech enhancement models.

cs.SD↗

arXiv2022

Semi-blind source separation using convolutive transfer function for nonlinear acoustic echo cancellation

The recently proposed semi-blind source separation (SBSS) method for nonlinear acoustic echo cancellation (NAEC) outperforms adaptive NAEC in attenuating the nonlinear acoustic echo. However, the multiplicative transfer function (MTF) approximation makes it unsuitable for real-time applications especially in highly reverberant environments, and the natural gradient makes it hard to balance well between fast convergence speed and stability. In this paper, we propose two more effective SBSS methods based on auxiliary-function-based independent vector analysis (AuxIVA) and independent low-rank matrix analysis (ILRMA). The convolutive transfer function (CTF) approximation is used instead of MTF so that a long impulse response can be modeled with a short latency. The optimization schemes used in AuxIVA and ILRMA are carefully regularized according to the constrained demixing matrix of NAEC. Experimental results validate significantly better echo cancellation performance of the proposed methods.

eess.AS↗

arXiv2022

A light-weight full-band speech enhancement model

Deep neural network based full-band speech enhancement systems face challenges of high demand of computational resources and imbalanced frequency distribution. In this paper, a light-weight full-band model is proposed with two dedicated strategies, i.e., a learnable spectral compression mapping for more effective high-band spectral information compression, and the utilization of the multi-head attention mechanism for more effective modeling of the global spectral pattern. Experiments validate the efficacy of the proposed strategies and show that the proposed model achieves competitive performance with only 0.89M parameters.

eess.AS↗

arXiv2022

A two-stage full-band speech enhancement model with effective spectral compression mapping

The direct expansion of deep neural network (DNN) based wide-band speech enhancement (SE) to full-band processing faces the challenge of low frequency resolution in low frequency range, which would highly likely lead to deteriorated performance of the model. In this paper, we propose a learnable spectral compression mapping (SCM) to effectively compress the high frequency components so that they can be processed in a more efficient manner. By doing so, the model can pay more attention to low and middle frequency range, where most of the speech power is concentrated. Instead of suppressing noise in a single network structure, we first estimate a spectral magnitude mask, converting the speech to a high signal-to-ratio (SNR) state, and then utilize a subsequent model to further optimize the real and imaginary mask of the pre-enhanced signal. We conduct comprehensive experiments to validate the efficacy of the proposed method.

cs.SD↗

arXiv2022

ED2LM: Encoder-Decoder to Language Model for Faster Document Re-ranking Inference

State-of-the-art neural models typically encode document-query pairs using cross-attention for re-ranking. To this end, models generally utilize an encoder-only (like BERT) paradigm or an encoder-decoder (like T5) approach. These paradigms, however, are not without flaws, i.e., running the model on all query-document pairs at inference-time incurs a significant computational cost. This paper proposes a new training and inference paradigm for re-ranking. We propose to finetune a pretrained encoder-decoder model using in the form of document to query generation. Subsequently, we show that this encoder-decoder architecture can be decomposed into a decoder-only language model during inference. This results in significant inference time speedups since the decoder-only architecture only needs to learn to interpret static encoder embeddings during inference. Our experiments show that this new paradigm achieves results that are comparable to the more expensive cross-attention ranking approaches while being up to 6.8X faster. We believe this work paves the way for more efficient neural rankers that leverage large pretrained models.

cs.CL↗

arXiv2022

PMAL: Open Set Recognition via Robust Prototype Mining

Open Set Recognition (OSR) has been an emerging topic. Besides recognizing predefined classes, the system needs to reject the unknowns. Prototype learning is a potential manner to handle the problem, as its ability to improve intra-class compactness of representations is much needed in discrimination between the known and the unknowns. In this work, we propose a novel Prototype Mining And Learning (PMAL) framework. It has a prototype mining mechanism before the phase of optimizing embedding space, explicitly considering two crucial properties, namely high-quality and diversity of the prototype set. Concretely, a set of high-quality candidates are firstly extracted from training samples based on data uncertainty learning, avoiding the interference from unexpected noise. Considering the multifarious appearance of objects even in a single category, a diversity-based strategy for prototype set filtering is proposed. Accordingly, the embedding space can be better optimized to discriminate therein the predefined classes and between known and unknowns. Extensive experiments verify the two good characteristics (i.e., high-quality and diversity) embraced in prototype mining, and show the remarkable performance of the proposed framework compared to state-of-the-arts.

cs.CV↗

arXiv2022

High-performance and Low-power Transistors Based on Anisotropic Monolayer $β$-TeO$_2$

Two-dimensional (2D) semiconductors offer a promising prospect for high-performance and energy-efficient devices especially in the sub-10 nm regime. Inspired by the successful fabrication of 2D $β$-TeO$_2$ and the high on/off ratio and high air-stability of fabricated field effect transistors (FETs) [Nat. Electron. 2021, 4, 277], we provide a comprehensive investigation of the electronic structure of monolayer $β$-TeO$_2$ and the device performance of sub-10 nm metal oxide semiconductors FETs (MOSFETs) based on this material. The anisotropic electronic structure of monolayer $β$-TeO$_2$ plays a critical role in the anisotropy of transport properties for MOSFETs. We show that the 5.2-nm gate-length n-type MOSFET holds an ultra-high on-state current exceeding 3700 μA/μm according to International Roadmap for Devices and Systems (IRDS) 2020 goals for high-performance devices, which is benefited by the highly anisotropic electron effective mass. Moreover, monolayer $β$-TeO$_2$ MOSFETs can fulfill the IRDS 2020 goals for both high-performance and low-power devices in terms of on-state current, sub-threshold swing, delay time, and power-delay product. This study unveils monolayer $β$-TeO$_2$ as a promising candidate for ultra-scaled devices in future nanoelectronics.

cond-mat.mes-hall↗

arXiv2022

Out-of-Domain Semantics to the Rescue! Zero-Shot Hybrid Retrieval Models

The pre-trained language model (eg, BERT) based deep retrieval models achieved superior performance over lexical retrieval models (eg, BM25) in many passage retrieval tasks. However, limited work has been done to generalize a deep retrieval model to other tasks and domains. In this work, we carefully select five datasets, including two in-domain datasets and three out-of-domain datasets with different levels of domain shift, and study the generalization of a deep model in a zero-shot setting. Our findings show that the performance of a deep retrieval model is significantly deteriorated when the target domain is very different from the source domain that the model was trained on. On the contrary, lexical models are more robust across domains. We thus propose a simple yet effective framework to integrate lexical and deep retrieval models. Our experiments demonstrate that these two models are complementary, even when the deep model is weaker in the out-of-domain setting. The hybrid model obtains an average of 20.4% relative gain over the deep retrieval model, and an average of 9.54% over the lexical model in three out-of-domain datasets.

cs.IR↗

arXiv2022

Interpretable learning of voltage for electrode design of multivalent metal-ion batteries

Deep learning (DL) has indeed emerged as a powerful tool for rapidly and accurately predicting materials properties from big data, such as the design of current commercial Li-ion batteries. However, its practical utility for multivalent metal-ion batteries (MIBs), the most promising future solution of large-scale energy storage, is limited due to the scarce MIB data availability and poor DL model interpretability. Here, we develop an interpretable DL model as an effective and accurate method for learning electrode voltages of multivalent MIBs (divalent magnesium, calcium, zinc, and trivalent aluminum) at small dataset limits (150~500). Using the experimental results as validation, our model is much more accurate than machine-learning models which usually are better than DL in the small dataset regime. Besides the high accuracy, our feature-engineering-free DL model is explainable, which automatically extracts the atom covalent radius as the most important feature for the voltage learning by visualizing vectors from the layers of the neural network. The presented model potentially accelerates the design and optimization of multivalent MIB materials with fewer data and less domain-knowledge restriction, and is implemented into a publicly available online tool kit in http://batteries.2dmatpedia.org/ for the battery community.

cond-mat.mtrl-sci↗

arXiv2021

Large Dual Encoders Are Generalizable Retrievers

It has been shown that dual encoders trained on one domain often fail to generalize to other domains for retrieval tasks. One widespread belief is that the bottleneck layer of a dual encoder, where the final score is simply a dot-product between a query vector and a passage vector, is too limited to make dual encoders an effective retrieval model for out-of-domain generalization. In this paper, we challenge this belief by scaling up the size of the dual encoder model {\em while keeping the bottleneck embedding size fixed.} With multi-stage training, surprisingly, scaling up the model size brings significant improvement on a variety of retrieval tasks, especially for out-of-domain generalization. Experimental results show that our dual encoders, \textbf{G}eneralizable \textbf{T}5-based dense \textbf{R}etrievers (GTR), outperform %ColBERT~\cite{khattab2020colbert} and existing sparse and dense retrievers on the BEIR dataset~\cite{thakur2021beir} significantly. Most surprisingly, our ablation study finds that GTR is very data efficient, as it only needs 10\% of MS Marco supervised data to achieve the best out-of-domain performance. All the GTR models are released at https://tfhub.dev/google/collections/gtr/1.

cs.IR↗

arXiv2021

Sliding Ferroelectric Tunnel Junctions

Very recently, ferroelectric polarization in staggered bilayer hexagonal boron nitride (BBN) and its novel sliding inversion mechanism was reported experimentally (Science 2021, 372, 1458; 2021, 372, 1462), which paves a new way to realize van der Waals (vdW) ferroelectric devices with new functionalities. Here, we develop vdW sliding ferroelectric tunnel junctions (FTJs) using the sliding ferroelectric BBN unit as ultrathin barriers and explore their transport properties with different ferroelectric states and metal contacts via the first principles. It is found that the electrode/BBN contact electric field quenches the ferroelectricity in the staggered BBN, resulting a very small tunnelling electroresistance (TER). Inserting high-mobility 2D materials between Au and BN can restore the BBN ferroelectricity, reaching a giant TER of ~10,000% in sliding FTJs. We finally investigate the metal-contact and thickness effect on the tunnelling property of sliding FTJs. The giant TER and multiple non-volatile resistance states in vdW sliding FTJs show the promising applications in voltage-controlled nano-memories with ultrahigh storage density.

cond-mat.mtrl-sci↗

arXiv2021

You Only Recognize Once: Towards Fast Video Text Spotting

Video text spotting is still an important research topic due to its various real-applications. Previous approaches usually fall into the four-staged pipeline: text detection in individual images, framewisely recognizing localized text regions, tracking text streams and generating final results with complicated post-processing skills, which might suffer from the huge computational cost as well as the interferences of low-quality text. In this paper, we propose a fast and robust video text spotting framework by only recognizing the localized text one-time instead of frame-wisely recognition. Specifically, we first obtain text regions in videos with a well-designed spatial-temporal detector. Then we concentrate on developing a novel text recommender for selecting the highest-quality text from text streams and only recognizing the selected ones. Here, the recommender assembles text tracking, quality scoring and recognition into an end-to-end trainable module, which not only avoids the interferences from low-quality text but also dramatically speeds up the video text spotting process. In addition, we collect a larger scale video text dataset (LSVTD) for promoting the video text spotting community, which contains 100 text videos from 22 different real-life scenarios. Extensive experiments on two public benchmarks show that our method greatly speeds up the recognition process averagely by 71 times compared with the frame-wise manner, and also achieves the remarkable state-of-the-art.

cs.CV↗

arXiv2021

TRIE: End-to-End Text Reading and Information Extraction for Document Understanding

Since real-world ubiquitous documents (e.g., invoices, tickets, resumes and leaflets) contain rich information, automatic document image understanding has become a hot topic. Most existing works decouple the problem into two separate tasks, (1) text reading for detecting and recognizing texts in images and (2) information extraction for analyzing and extracting key elements from previously extracted plain text. However, they mainly focus on improving information extraction task, while neglecting the fact that text reading and information extraction are mutually correlated. In this paper, we propose a unified end-to-end text reading and information extraction network, where the two tasks can reinforce each other. Specifically, the multimodal visual and textual features of text reading are fused for information extraction and in turn, the semantics in information extraction contribute to the optimization of text reading. On three real-world datasets with diverse document images (from fixed layout to variable layout, from structured text to semi-structured text), our proposed method significantly outperforms the state-of-the-art methods in both efficiency and accuracy.

cs.CV↗

arXiv2021

SN 2021csp -- the explosion of a stripped envelope star within a H and He-poor circumstellar medium

We present observations of SN 2021csp, a unique supernova (SN) which displays evidence for interaction with H- and He- poor circumstellar material (CSM) at early times. Using high-cadence spectroscopy taken over the first week after explosion, we show that the spectra of SN 2021csp are dominated by C III lines with a velocity of 1800 km s$^{-1}$. We associate this emission with CSM lost by the progenitor prior to explosion. Subsequently, the SN displays narrow He lines before metamorphosing into a broad-lined Type Ic SN. We model the bolometric light curve of SN 2021csp, and show that it is consistent with the energetic ($4\times10^{51}$ erg) explosion of a stripped star, producing 0.4 M$_\odot$ of 56Ni within a $\sim$1 M$_\odot$ shell of CSM extending out to 400 R$_\odot$.

astro-ph.SR↗

arXiv2021

ICDAR 2021 Competition on Scene Video Text Spotting

Scene video text spotting (SVTS) is a very important research topic because of many real-life applications. However, only a little effort has put to spotting scene video text, in contrast to massive studies of scene text spotting in static images. Due to various environmental interferences like motion blur, spotting scene video text becomes very challenging. To promote this research area, this competition introduces a new challenge dataset containing 129 video clips from 21 natural scenarios in full annotations. The competition containts three tasks, that is, video text detection (Task 1), video text tracking (Task 2) and end-to-end video text spotting (Task3). During the competition period (opened on 1st March, 2021 and closed on 11th April, 2021), a total of 24 teams participated in the three proposed tasks with 46 valid submissions, respectively. This paper includes dataset descriptions, task definitions, evaluation protocols and results summaries of the ICDAR 2021 on SVTS competition. Thanks to the healthy number of teams as well as submissions, we consider that the SVTS competition has been successfully held, drawing much attention from the community and promoting the field research and its development.

cs.CV↗