Search arXiv⌕ Search

arXiv subjects

Yu Zhang

Publications and source records attributed to Yu Zhang.

At least 1,045 records · Page 58Linked to original sources
arXiv2023

Edgeformers: Graph-Empowered Transformers for Representation Learning on Textual-Edge Networks

Edges in many real-world social/information networks are associated with rich text information (e.g., user-user communications or user-product reviews). However, mainstream network representation learning models focus on propagating and aggregating node attributes, lacking specific designs to utilize text semantics on edges. While there exist edge-aware graph neural networks, they directly initialize edge attributes as a feature vector, which cannot fully capture the contextualized text semantics of edges. In this paper, we propose Edgeformers, a framework built upon graph-enhanced Transformers, to perform edge and node representation learning by modeling texts on edges in a contextualized way. Specifically, in edge representation learning, we inject network information into each Transformer layer when encoding edge texts; in node representation learning, we aggregate edge representations through an attention mechanism within each node's ego-graph. On five public datasets from three different domains, Edgeformers consistently outperform state-of-the-art baselines in edge classification and link prediction, demonstrating the efficacy in learning edge and node representations, respectively.

cs.LG↗
arXiv2023

Directional carrier transport in micrometer-thick gallium oxide films for high-performance deep-ultraviolet photodetection

Incorporating emerging ultrawide bandgap semiconductors with a metal-semiconductor-metal (MSM) architecture is highly desired for deep-ultraviolet (DUV) photodetection. However, synthesis-induced defects in semiconductors complicate the rational design of MSM DUV photodetectors due to their dual role as carrier donors and trap centers, leading to a commonly observed trade-off between responsivity and response time. Here, we demonstrate a simultaneous improvement of these two parameters in ε-Ga2O3 MSM photodetectors by establishing a low-defect diffusion barrier for directional carrier transport. Specifically, using a micrometer thickness far exceeding its effective light absorption depth, the ε-Ga2O3 MSM photodetector achieves over 18-fold enhancement of responsivity and simultaneous reduction of the response time, which exhibits a state-of-the-art photo-to-dark current ratio near 10^8, a superior responsivity of >1300 A/W, an ultrahigh detectivity of >10^16 Jones and a decay time of 123 ms. Combined depth-profile spectroscopic and microscopic analysis reveals the existence of a broad defective region near the lattice-mismatched interface followed by a more defect-free dark region, while the latter one serves as a diffusion barrier to assist frontward carrier transport for substantially enhancing the photodetector performance. This work reveals the critical role of the semiconductor defect profile in tuning carrier transport for fabricating high-performance MSM DUV photodetectors.

cond-mat.mtrl-sci↗
arXiv2023

Modular Hybrid Autoregressive Transducer

Text-only adaptation of a transducer model remains challenging for end-to-end speech recognition since the transducer has no clearly separated acoustic model (AM), language model (LM) or blank model. In this work, we propose a modular hybrid autoregressive transducer (MHAT) that has structurally separated label and blank decoders to predict label and blank distributions, respectively, along with a shared acoustic encoder. The encoder and label decoder outputs are directly projected to AM and internal LM scores and then added to compute label posteriors. We train MHAT with an internal LM loss and a HAT loss to ensure that its internal LM becomes a standalone neural LM that can be effectively adapted to text. Moreover, text adaptation of MHAT fosters a much better LM fusion than internal LM subtraction-based methods. On Google's large-scale production data, a multi-domain MHAT adapted with 100B sentences achieves relative WER reductions of up to 12.4% without LM fusion and 21.5% with LM fusion from 400K-hour trained HAT.

cs.CL↗
arXiv2023

Network-Assisted Full-Duplex Cell-Free mmWave Massive MIMO Systems with DAC Quantization and Fronthaul Compression

In this paper, we investigate network-assisted full-duplex (NAFD) cell-free millimeter-wave (mmWave) massive multiple-input multiple-output (MIMO) systems with digital-to-analog converter (DAC) quantization and fronthaul compression. We propose to maximize the weighted uplink and downlink sum rate by jointly optimizing the power allocation of both the transmitting remote antenna units (T-RAUs) and uplink users and the variances of the downlink and uplink fronthaul compression noises. To deal with this challenging problem, we further apply a successive convex approximation (SCA) method to handle the non-convex bidirectional limited-capacity fronthaul constraints. The simulation results verify the convergence of the proposed SCA-based algorithm and analyze the impact of fronthaul capacity and DAC quantization on the spectral efficiency of the NAFD cell-free mmWave massive MIMO systems. Moreover, some insightful conclusions are obtained through the comparisons of spectral efficiency, which shows that NAFD achieves better performance gains than co-time co-frequency full-duplex cloud radio access network (CCFD C-RAN) in the cases of practical limited-resolution DACs. Specifically, their performance gaps with 8-bit DAC quantization are larger than that with 1-bit DAC quantization, which attains a 5.5-fold improvement.

cs.IT↗
arXiv2023

Massively Multilingual Shallow Fusion with Large Language Models

While large language models (LLM) have made impressive progress in natural language processing, it remains unclear how to utilize them in improving automatic speech recognition (ASR). In this work, we propose to train a single multilingual language model (LM) for shallow fusion in multiple languages. We push the limits of the multilingual LM to cover up to 84 languages by scaling up using a mixture-of-experts LLM, i.e., generalist language model (GLaM). When the number of experts increases, GLaM dynamically selects only two at each decoding step to keep the inference computation roughly constant. We then apply GLaM to a multilingual shallow fusion task based on a state-of-the-art end-to-end model. Compared to a dense LM of similar computation during inference, GLaM reduces the WER of an English long-tail test set by 4.4% relative. In a multilingual shallow fusion task, GLaM improves 41 out of 50 languages with an average relative WER reduction of 3.85%, and a maximum reduction of 10%. Compared to the baseline model, GLaM achieves an average WER reduction of 5.53% over 43 languages.

cs.CL↗
arXiv2023

JEIT: Joint End-to-End Model and Internal Language Model Training for Speech Recognition

We propose JEIT, a joint end-to-end (E2E) model and internal language model (ILM) training method to inject large-scale unpaired text into ILM during E2E training which improves rare-word speech recognition. With JEIT, the E2E model computes an E2E loss on audio-transcript pairs while its ILM estimates a cross-entropy loss on unpaired text. The E2E model is trained to minimize a weighted sum of E2E and ILM losses. During JEIT, ILM absorbs knowledge from unpaired text while the E2E training serves as regularization. Unlike ILM adaptation methods, JEIT does not require a separate adaptation step and avoids the need for Kullback-Leibler divergence regularization of ILM. We also show that modular hybrid autoregressive transducer (MHAT) performs better than HAT in the JEIT framework, and is much more robust than HAT during ILM adaptation. To push the limit of unpaired text injection, we further propose a combined JEIT and JOIST training (CJJT) that benefits from modality matching, encoder text injection and ILM training. Both JEIT and CJJT can foster a more effective LM fusion. With 100B unpaired sentences, JEIT/CJJT improves rare-word recognition accuracy by up to 16.4% over a model trained without unpaired text.

eess.AS↗
arXiv2023

Hidden Euclidean dynamical symmetry in the U(n+1) vibron model

Based on the boson realization of the Euclidean algebras, it is found that the E($n$) dynamical symmetry (DS) may emerge at the critical point of the U($n$)-SO($n+1$) quantum phase transition. To justify this finding, we provide a detailed analysis of the critical dynamics in the U($n+1$) vibron model in both quantal and classical ways. It is further shown that the low-lying structure of $^{82}$Kr may serve as an excellent empirical realization of the E(5) DS in experiments.

nucl-th↗
arXiv2023

Quadrupole and octupole states in $^{152}$Sm using the proton-neutron interacting boson model

A scheme of solving the proton-neutron interacting boson model (IBM-2) in terms of the SU(3) basis is introduced, by which the IBM-2 coupled with an octupole boson is applied to describe the low-energy structure of the critical point nucleus, $^{152}$Sm. The results indicate that the spectral properties of both the positive-parity bands and negative-parity bands in this nucleus can be well captured by the IBM-2 calculations through a simple Hamiltonian, thus providing an example of the IBM-2 in a unified description of quadrupole and octupole states in a transitional system. In addition, a statistical analysis of the low-spin states in the model is also provided.

nucl-th↗
arXiv2023

The Effect of Metadata on Scientific Literature Tagging: A Cross-Field Cross-Model Study

Due to the exponential growth of scientific publications on the Web, there is a pressing need to tag each paper with fine-grained topics so that researchers can track their interested fields of study rather than drowning in the whole literature. Scientific literature tagging is beyond a pure multi-label text classification task because papers on the Web are prevalently accompanied by metadata information such as venues, authors, and references, which may serve as additional signals to infer relevant tags. Although there have been studies making use of metadata in academic paper classification, their focus is often restricted to one or two scientific fields (e.g., computer science and biomedicine) and to one specific model. In this work, we systematically study the effect of metadata on scientific literature tagging across 19 fields. We select three representative multi-label classifiers (i.e., a bag-of-words model, a sequence-based model, and a pre-trained language model) and explore their performance change in scientific literature tagging when metadata are fed to the classifiers as additional features. We observe some ubiquitous patterns of metadata's effects across all fields (e.g., venues are consistently beneficial to paper tagging in almost all cases), as well as some unique patterns in fields other than computer science and biomedicine, which are not explored in previous studies.

cs.DL↗
arXiv2023

TrajMatch: Towards Automatic Spatio-temporal Calibration for Roadside LiDARs through Trajectory Matching

Recently, it has become popular to deploy sensors such as LiDARs on the roadside to monitor the passing traffic and assist autonomous vehicle perception. Unlike autonomous vehicle systems, roadside sensors are usually affiliated with different subsystems and lack synchronization both in time and space. Calibration is a key technology which allows the central server to fuse the data generated by different location infrastructures, which can deliver improve the sensing range and detection robustness. Unfortunately, existing calibration algorithms often assume that the LiDARs are significantly overlapped or that the temporal calibration is already achieved. Since these assumptions do not always hold in the real world, the calibration results from the existing algorithms are often unsatisfactory and always need human involvement, which brings high labor costs. In this paper, we propose TrajMatch -- the first system that can automatically calibrate for roadside LiDARs in both time and space. The main idea is to automatically calibrate the sensors based on the result of the detection/tracking task instead of extracting special features. More deeply, we propose a mechanism for evaluating calibration parameters that is consistent with our algorithm, and we demonstrate the effectiveness of this scheme experimentally, which can also be used to guide parameter iterations for multiple calibration. Finally, to evaluate the performance of TrajMatch , we collect two dataset, one simulated dataset LiDARnet-sim 1.0 and a real-world dataset. Experiment results show that TrajMatch can achieve a spatial calibration error of less than 10cm and a temporal calibration error of less than 1.5ms.

cs.RO↗
arXiv2023

Efficient Domain Adaptation for Speech Foundation Models

Foundation models (FMs), that are trained on broad data at scale and are adaptable to a wide range of downstream tasks, have brought large interest in the research community. Benefiting from the diverse data sources such as different modalities, languages and application domains, foundation models have demonstrated strong generalization and knowledge transfer capabilities. In this paper, we present a pioneering study towards building an efficient solution for FM-based speech recognition systems. We adopt the recently developed self-supervised BEST-RQ for pretraining, and propose the joint finetuning with both source and unsupervised target domain data using JUST Hydra. The FM encoder adapter and decoder are then finetuned to the target domain with a small amount of supervised in-domain data. On a large-scale YouTube and Voice Search task, our method is shown to be both data and model parameter efficient. It achieves the same quality with only 21.6M supervised in-domain data and 130.8M finetuned parameters, compared to the 731.1M model trained from scratch on additional 300M supervised in-domain data.

cs.CL↗
arXiv2023

A Digital Twin Assisted Framework for Interference Nulling in Millimeter Wave MIMO Systems

Millimeter wave (mmWave) and terahertz MIMO systems rely on pre-defined beamforming codebooks for both initial access and data transmission. However, most of the existing codebooks adopt pre-defined beams that focus mainly on improving the gain of their target users, without taking interference into account, which could incur critical performance degradation in dense networks. To address this problem, in this paper, we propose a sample-efficient digital twin-assisted beam pattern design framework that learns how to form the beam pattern to reject the signals from the interfering directions. The proposed approach does not require any explicit channel knowledge or any coordination with the interferers. The adoption of the digital twin improves the sample efficiency by better leveraging the underlying signal relationship and by incorporating a demand-based data acquisition strategy. Simulation results show that the developed signal model-based learning framework can significantly reduce the actual interaction with the radio environment (i.e., the number of measurements) compared to the model-unaware design, leading to a more practical and efficient interference-aware beam design approach.

cs.IT↗
arXiv2023

Improving Responsiveness to Robots for Tacit Human-Robot Interaction via Implicit and Naturalistic Team Status Projection

Fluent human-human teaming is often characterized by tacit interaction without explicit communication. This is because explicit communication, such as language utterances and gestures, are inherently interruptive. On the other hand, tacit interaction requires team situation awareness (TSA) to facilitate, which often relies on explicit communication to maintain, creating a paradox. In this paper, we consider implicit and naturalistic team status projection for tacit human-robot interaction. Implicitness minimizes interruption while naturalness reduces cognitive demand, and they together improve responsiveness to robots. We introduce a novel process for such Team status Projection via virtual Shadows, or TPS. We compare our method with two baselines that use explicit projection for maintaining TSA. Results via human factors studies demonstrate that TPS provides a more fluent human-robot interaction experience by significantly improving human responsiveness to robots in tacit teaming scenarios, which suggests better TSA. Participants acknowledged robots implementing TPS as more acceptable as a teammate and favorable. Simultaneously, we demonstrate that TPS is comparable to, and sometimes better than, the best-performing baseline in maintaining accurate TSA

cs.RO↗
arXiv2023

From English to More Languages: Parameter-Efficient Model Reprogramming for Cross-Lingual Speech Recognition

In this work, we propose a new parameter-efficient learning framework based on neural model reprogramming for cross-lingual speech recognition, which can \textbf{re-purpose} well-trained English automatic speech recognition (ASR) models to recognize the other languages. We design different auxiliary neural architectures focusing on learnable pre-trained feature enhancement that, for the first time, empowers model reprogramming on ASR. Specifically, we investigate how to select trainable components (i.e., encoder) of a conformer-based RNN-Transducer, as a frozen pre-trained backbone. Experiments on a seven-language multilingual LibriSpeech speech (MLS) task show that model reprogramming only requires 4.2% (11M out of 270M) to 6.8% (45M out of 660M) of its original trainable parameters from a full ASR model to perform competitive results in a range of 11.9% to 8.1% WER averaged across different languages. In addition, we discover different setups to make large-scale pre-trained ASR succeed in both monolingual and multilingual speech recognition. Our methods outperform existing ASR tuning architectures and their extension with self-supervised losses (e.g., w2v-bert) in terms of lower WER and better training efficiency.

cs.SD↗
arXiv2023

Super-Resolution Harmonic Retrieval of Non-Circular Signals

This paper proposes a super-resolution harmonic retrieval method for uncorrelated strictly non-circular signals, whose covariance and pseudo-covariance present Toeplitz and Hankel structures, respectively. Accordingly, the augmented covariance matrix constructed by the covariance and pseudo-covariance matrices is not only low rank but also jointly Toeplitz-Hankel structured. To efficiently exploit such a desired structure for high estimation accuracy, we develop a low-rank Toeplitz-Hankel covariance reconstruction (LRTHCR) solution employed over the augmented covariance matrix. Further, we design a fitting error constraint to flexibly implement the LRTHCR algorithm without knowing the noise statistics. In addition, performance analysis is provided for the proposed LRTHCR in practical settings. Simulation results reveal that the LRTHCR outperforms the benchmark methods in terms of lower estimation errors.

eess.SP↗
arXiv2023

Quantum simulation of molecular response properties

Accurate modeling of the response of molecular systems to an external electromagnetic field is challenging on classical computers, especially in the regime of strong electronic correlation. In this paper, we develop a quantum linear response (qLR) theory to calculate molecular response properties on near-term quantum computers. Inspired by the recently developed variants of the quantum counterpart of equation of motion (qEOM) theory, the qLR formalism employs "killer condition" satisfying excitation operator manifolds that offers a number of theoretical advantages along with reduced quantum resource requirements. We also used the qEOM framework in this work to calculate state-specific response properties. Further, through noise-less quantum simulations, we show that response properties calculated using the qLR approach are more accurate than the ones obtained from the classical coupled-cluster based linear response models due to the improved quality of the ground-state wavefunction obtained using the ADAPT-VQE algorithm.

quant-ph↗
arXiv2023

MedSegDiff: Medical Image Segmentation with Diffusion Probabilistic Model

Diffusion probabilistic model (DPM) recently becomes one of the hottest topic in computer vision. Its image generation application such as Imagen, Latent Diffusion Models and Stable Diffusion have shown impressive generation capabilities, which aroused extensive discussion in the community. Many recent studies also found it is useful in many other vision tasks, like image deblurring, super-resolution and anomaly detection. Inspired by the success of DPM, we propose the first DPM based model toward general medical image segmentation tasks, which we named MedSegDiff. In order to enhance the step-wise regional attention in DPM for the medical image segmentation, we propose dynamic conditional encoding, which establishes the state-adaptive conditions for each sampling step. We further propose Feature Frequency Parser (FF-Parser), to eliminate the negative effect of high-frequency noise component in this process. We verify MedSegDiff on three medical segmentation tasks with different image modalities, which are optic cup segmentation over fundus images, brain tumor segmentation over MRI images and thyroid nodule segmentation over ultrasound images. The experimental results show that MedSegDiff outperforms state-of-the-art (SOTA) methods with considerable performance gap, indicating the generalization and effectiveness of the proposed model. Our code is released at https://github.com/WuJunde/MedSegDiff.

cs.CV↗
arXiv2023

Tac2Structure: Object Surface Reconstruction Only through Multi Times Touch

Inspired by humans' ability to perceive the surface texture of unfamiliar objects without relying on vision, the sense of touch can play a crucial role in robots exploring the environment, particularly in scenes where vision is difficult to apply, or occlusion is inevitable. Existing tactile surface reconstruction methods rely on external sensors or have strong prior assumptions, making the operation complex and limiting their application scenarios. This paper presents a framework for low-drift surface reconstruction through multiple tactile measurements, Tac2Structure. Compared with existing algorithms, the proposed method uses only a new vision-based tactile sensor without relying on external devices. Aiming at the difficulty that reconstruction accuracy is easily affected by the pressure at contact, we propose a correction algorithm to adapt it. The proposed method also reduces the accumulative errors that occur easily during global object surface reconstruction. Multi-frame tactile measurements can accurately reconstruct object surfaces by jointly using the point cloud registration algorithm, loop-closure detection algorithm based on deep learning, and pose graph optimization algorithm. Experiments verify that Tac2Structure can achieve millimeter-level accuracy in reconstructing the surface of objects, providing accurate tactile information for the robot to perceive the surrounding environment.

cs.RO↗