arXiv subjects

Jing Shi

Publications and source records attributed to Jing Shi.

At least 73 records · Page 4Linked to original sources

arXiv2023

InstantBooth: Personalized Text-to-Image Generation without Test-Time Finetuning

Recent advances in personalized image generation allow a pre-trained text-to-image model to learn a new concept from a set of images. However, existing personalization approaches usually require heavy test-time finetuning for each concept, which is time-consuming and difficult to scale. We propose InstantBooth, a novel approach built upon pre-trained text-to-image models that enables instant text-guided image personalization without any test-time finetuning. We achieve this with several major components. First, we learn the general concept of the input images by converting them to a textual token with a learnable image encoder. Second, to keep the fine details of the identity, we learn rich visual feature representation by introducing a few adapter layers to the pre-trained model. We train our components only on text-image pairs without using paired images of the same concept. Compared to test-time finetuning-based methods like DreamBooth and Textual-Inversion, our model can generate competitive results on unseen concepts concerning language-image alignment, image fidelity, and identity preservation while being 100 times faster.

cs.CV↗

arXiv2023

Matching-based Term Semantics Pre-training for Spoken Patient Query Understanding

Medical Slot Filling (MSF) task aims to convert medical queries into structured information, playing an essential role in diagnosis dialogue systems. However, the lack of sufficient term semantics learning makes existing approaches hard to capture semantically identical but colloquial expressions of terms in medical conversations. In this work, we formalize MSF into a matching problem and propose a Term Semantics Pre-trained Matching Network (TSPMN) that takes both terms and queries as input to model their semantic interaction. To learn term semantics better, we further design two self-supervised objectives, including Contrastive Term Discrimination (CTD) and Matching-based Mask Term Modeling (MMTM). CTD determines whether it is the masked term in the dialogue for each given term, while MMTM directly predicts the masked ones. Experimental results on two Chinese benchmarks show that TSPMN outperforms strong baselines, especially in few-shot settings.

cs.CL↗

arXiv2023

In-situ electrical and thermal transport properties of FeySe1-xTex films with ionic liquid gating

We combine in-situ electrical transport and Seebeck coefficient measurements with the ionic liquid gating technique to investigate superconductivity and the normal state of FeySe1-xTex (FST) films. We find that the pristine FST films feature a non-Fermi liquid temperature dependence of the Seebeck coefficient, i.e., S/T ~ AS lnT, and AS is strongly correlated with the superconducting transition temperature (Tc). Ionic liquid gating significantly raises Tc of FST films, for which the Seebeck coefficient displays a novel scaling behavior and retains the logarithmic temperature dependence. Moreover, a quantitative relationship between the slope of T-linear resistivity (A\r{ho}) and Tc for gated films is observed, i.e., (A\r{ho})1/2 ~ Tc, consistent with previous reports on cuprates and FeSe. The scaling behaviors of AS and A\r{ho} point to a spin-fluctuation-associated transport mechanism in gated FeySe1-xTex superconductors.

cond-mat.supr-con↗

arXiv2022

Anomaly Crossing: New Horizons for Video Anomaly Detection as Cross-domain Few-shot Learning

Video anomaly detection aims to identify abnormal events that occurred in videos. Since anomalous events are relatively rare, it is not feasible to collect a balanced dataset and train a binary classifier to solve the task. Thus, most previous approaches learn only from normal videos using unsupervised or semi-supervised methods. Obviously, they are limited in capturing and utilizing discriminative abnormal characteristics, which leads to compromised anomaly detection performance. In this paper, to address this issue, we propose a new learning paradigm by making full use of both normal and abnormal videos for video anomaly detection. In particular, we formulate a new learning task: cross-domain few-shot anomaly detection, which can transfer knowledge learned from numerous videos in the source domain to help solve few-shot abnormality detection in the target domain. Concretely, we leverage self-supervised training on the target normal videos to reduce the domain gap and devise a meta context perception module to explore the video context of the event in the few-shot setting. Our experiments show that our method significantly outperforms baseline methods on DoTA and UCF-Crime datasets, and the new task contributes to a more practical training paradigm for anomaly detection.

cs.CV↗

arXiv2022

Dynamic Event-Triggered Discrete-Time Linear Time-Varying System with Privacy-Preservation

This paper focuses on discrete-time wireless sensor networks with privacy-preservation. In practical applications, information exchange between sensors is subject to attacks. For the information leakage caused by the attack during the information transmission process, privacy-preservation is introduced for system states. To make communication resources more effectively utilized, a dynamic event-triggered set-membership estimator is designed. Moreover, the privacy of the system is analyzed to ensure the security of the real data. As a result, the set-membership estimator with differential privacy is analyzed using recursive convex optimization. Then the steady-state performance of the system is studied. Finally, one example is presented to demonstrate the feasibility of the proposed distributed filter containing privacy-preserving analysis.

eess.SY↗

arXiv2022

VLP: A Survey on Vision-Language Pre-training

In the past few years, the emergence of pre-training models has brought uni-modal fields such as computer vision (CV) and natural language processing (NLP) to a new era. Substantial works have shown they are beneficial for downstream uni-modal tasks and avoid training a new model from scratch. So can such pre-trained models be applied to multi-modal tasks? Researchers have explored this problem and made significant progress. This paper surveys recent advances and new frontiers in vision-language pre-training (VLP), including image-text and video-text pre-training. To give readers a better overall grasp of VLP, we first review its recent advances from five aspects: feature extraction, model architecture, pre-training objectives, pre-training datasets, and downstream tasks. Then, we summarize the specific VLP models in detail. Finally, we discuss the new frontiers in VLP. To the best of our knowledge, this is the first survey focused on VLP. We hope that this survey can shed light on future research in the VLP field.

cs.CV↗

arXiv2022

Discretization and Re-synthesis: an alternative method to solve the Cocktail Party Problem

Deep learning based models have significantly improved the performance of speech separation with input mixtures like the cocktail party. Prominent methods (e.g., frequency-domain and time-domain speech separation) usually build regression models to predict the ground-truth speech from the mixture, using the masking-based design and the signal-level loss criterion (e.g., MSE or SI-SNR). This study demonstrates, for the first time, that the synthesis-based approach can also perform well on this problem, with great flexibility and strong potential. Specifically, we propose a novel speech separation/enhancement model based on the recognition of discrete symbols, and convert the paradigm of the speech separation/enhancement related tasks from regression to classification. By utilizing the synthesis model with the input of discrete symbols, after the prediction of discrete symbol sequence, each target speech could be re-synthesized. Evaluation results based on the WSJ0-2mix and VCTK-noisy corpora in various settings show that our proposed method can steadily synthesize the separated speech with high speech quality and without any interference, which is difficult to avoid in regression-based methods. In addition, with negligible loss of listening quality, the speaker conversion of enhanced/separated speech could be easily realized through our method.

cs.SD↗

arXiv2021

Wide Field Imaging of van der Waals Ferromagnet Fe3GeTe2 by Spin Defects in Hexagonal Boron Nitride

Emergent color centers with accessible spins hosted by van der Waals materials have attracted substantial interest in recent years due to their significant potential for implementing transformative quantum sensing technologies. Hexagonal boron nitride (hBN) is naturally relevant in this context due to its remarkable ease of integration into devices consisting of low-dimensional materials. Taking advantage of boron vacancy spin defects in hBN, we report nanoscale quantum imaging of low-dimensional ferromagnetism sustained in Fe3GeTe2/hBN van der Waals heterostructures. Exploiting spin relaxometry methods, we have further observed spatially varying magnetic fluctuations in the exfoliated Fe3GeTe2 flake, whose magnitude reaches a peak value around the Curie temperature. Our results demonstrate the capability of spin defects in hBN of investigating local magnetic properties of layered materials in an accessible and precise way, which can be extended readily to a broad range of miniaturized van der Waals heterostructure systems.

cond-mat.mtrl-sci↗

arXiv2021

Role of Magnon-Magnon Scattering in Magnon Polaron Spin Seebeck Effect

The spin Seebeck effect (SSE) signal of magnon polarons in bulk-Y3Fe5O12 (YIG)/Pt heterostructures is found to drastically change as a function of temperature. It appears as a dip in the total SSE signal at low temperatures, but as the temperature increases, the dip gradually decreases before turning to a peak. We attribute the observed dip-to-peak transition to the rapid rise of the four-magnon scattering rate. Our analysis provides important insights into the microscopic origin of the hybridized excitations and the overall temperature dependence of the SSE anomalies.

cond-mat.mtrl-sci↗

arXiv2021

SpaceEdit: Learning a Unified Editing Space for Open-Domain Image Editing

Recently, large pretrained models (e.g., BERT, StyleGAN, CLIP) have shown great knowledge transfer and generalization capability on various downstream tasks within their domains. Inspired by these efforts, in this paper we propose a unified model for open-domain image editing focusing on color and tone adjustment of open-domain images while keeping their original content and structure. Our model learns a unified editing space that is more semantic, intuitive, and easy to manipulate than the operation space (e.g., contrast, brightness, color curve) used in many existing photo editing softwares. Our model belongs to the image-to-image translation framework which consists of an image encoder and decoder, and is trained on pairs of before- and after-images to produce multimodal outputs. We show that by inverting image pairs into latent codes of the learned editing space, our model can be leveraged for various downstream editing tasks such as language-guided image editing, personalized editing, editing-style clustering, retrieval, etc. We extensively study the unique properties of the editing space in experiments and demonstrate superior performance on the aforementioned tasks.

cs.CV↗

arXiv2021

First and second order magnetic anisotropy and damping of europium iron garnet under high strain

Understanding and tailoring static and dynamic properties of magnetic insulator thin films is important for spintronic device applications. Here, we grow atomically flat epitaxial europium iron garnet (EuIG) thin films by pulsed laser deposition on (111)-oriented garnet substrates with a range of lattice parameters. By controlling the lattice mismatch between EuIG and the substrates, we tune the strain in EuIG films from compressive to tensile regime, which is characterized by X-ray diffraction. Using ferromagnetic resonance, we find that in addition to the first-order perpendicular magnetic anisotropy which depends linearly on the strain, there is a significant second-order one that has a quadratic strain dependence. Inhomogeneous linewidth of the ferromagnetic resonance increases notably with increasing strain, while the Gilbert damping parameter remains nearly constant (~ 2x10^-2). These results provide valuable insight into the spin dynamics in ferrimagnetic insulators and useful guidance for material synthesis and engineering of next-generation spintronics applications.

cond-mat.mtrl-sci↗

arXiv2021

Closing the Gap Between Time-Domain Multi-Channel Speech Enhancement on Real and Simulation Conditions

The deep learning based time-domain models, e.g. Conv-TasNet, have shown great potential in both single-channel and multi-channel speech enhancement. However, many experiments on the time-domain speech enhancement model are done in simulated conditions, and it is not well studied whether the good performance can generalize to real-world scenarios. In this paper, we aim to provide an insightful investigation of applying multi-channel Conv-TasNet based speech enhancement to both simulation and real data. Our preliminary experiments show a large performance gap between the two conditions in terms of the ASR performance. Several approaches are applied to close this gap, including the integration of multi-channel Conv-TasNet into the beamforming model with various strategies, and the joint training of speech enhancement and speech recognition models. Our experiments on the CHiME-4 corpus show that our proposed approaches can greatly reduce the speech recognition performance discrepancy between simulation and real data, while preserving the strong speech enhancement capability in the frontend.

eess.AS↗

arXiv2021

An Exploration of Self-Supervised Pretrained Representations for End-to-End Speech Recognition

Self-supervised pretraining on speech data has achieved a lot of progress. High-fidelity representation of the speech signal is learned from a lot of untranscribed data and shows promising performance. Recently, there are several works focusing on evaluating the quality of self-supervised pretrained representations on various tasks without domain restriction, e.g. SUPERB. However, such evaluations do not provide a comprehensive comparison among many ASR benchmark corpora. In this paper, we focus on the general applications of pretrained speech representations, on advanced end-to-end automatic speech recognition (E2E-ASR) models. We select several pretrained speech representations and present the experimental results on various open-source and publicly available corpora for E2E-ASR. Without any modification of the back-end model architectures or training strategy, some of the experiments with pretrained representations, e.g., WSJ, WSJ0-2mix with HuBERT, reach or outperform current state-of-the-art (SOTA) recognition performance. Moreover, we further explore more scenarios for whether the pretraining representations are effective, such as the cross-language or overlapped speech. The scripts, configuratons and the trained models have been released in ESPnet to let the community reproduce our experiments and improve them.

cs.CL↗

arXiv2021

Learning to Generate Scene Graph from Natural Language Supervision

Learning from image-text data has demonstrated recent success for many recognition tasks, yet is currently limited to visual features or individual visual concepts such as objects. In this paper, we propose one of the first methods that learn from image-sentence pairs to extract a graphical representation of localized objects and their relationships within an image, known as scene graph. To bridge the gap between images and texts, we leverage an off-the-shelf object detector to identify and localize object instances, match labels of detected regions to concepts parsed from captions, and thus create "pseudo" labels for learning scene graph. Further, we design a Transformer-based model to predict these "pseudo" labels via a masked token prediction task. Learning from only image-sentence pairs, our model achieves 30% relative gain over a latest method trained with human-annotated unlocalized scene graphs. Our model also shows strong results for weakly and fully supervised scene graph generation. In addition, we explore an open-vocabulary setting for detecting scene graphs, and present the first result for open-set scene graph generation. Our code is available at https://github.com/YiwuZhong/SGG_from_NLS.

cs.CV↗

arXiv2021

Ultrafast Measurements of the Interfacial Spin Seebeck Effect in Au and Rare-Earth Iron Garnet Bilayers

We investigate picosecond spin-currents across Au/iron-garnet interfaces in response to ultrafast laser heating of the electrons in the Au film. In the picoseconds after optical heating, interfacial spin currents occur due to an interfacial temperature difference between electrons in the metal and magnons in the insulator. We report measurements of this interfacial longitudinal spin Seebeck effect between Au and rare-earth iron-garnet insulators, i.e. RE$_3$ Fe$_5$O$_{12}$, where RE is Y, Eu, Tb, Tm. We use time domain thermoreflectance (TDTR) measurements to characterize the thermal response of the bilayer to ultrafast optical heating. We use time-resolved magneto-optic Kerr effect (TR-MOKE) measurements of the Au layer to measure the time-evolution of spin accumulation in the Au film. We observe a spin Seebeck effect between Au/TmIG that is three times larger than for an Au/YIG bilayer. The interfacial thermal conductance between electrons in the Au and magnons in the TmIG layer is ~ 3 $\frac{MW}{m^2 K}$.

cond-mat.mtrl-sci↗

arXiv2021

Learning by Planning: Language-Guided Global Image Editing

Recently, language-guided global image editing draws increasing attention with growing application potentials. However, previous GAN-based methods are not only confined to domain-specific, low-resolution data but also lacking in interpretability. To overcome the collective difficulties, we develop a text-to-operation model to map the vague editing language request into a series of editing operations, e.g., change contrast, brightness, and saturation. Each operation is interpretable and differentiable. Furthermore, the only supervision in the task is the target image, which is insufficient for a stable training of sequential decisions. Hence, we propose a novel operation planning algorithm to generate possible editing sequences from the target image as pseudo ground truth. Comparison experiments on the newly collected MA5k-Req dataset and GIER dataset show the advantages of our methods. Code is available at https://jshi31.github.io/T2ONet.

cs.CV↗

arXiv2021

Learning Continuous-Time Dynamics by Stochastic Differential Networks

Learning continuous-time stochastic dynamics is a fundamental and essential problem in modeling sporadic time series, whose observations are irregular and sparse in both time and dimension. For a given system whose latent states and observed data are high-dimensional, it is generally impossible to derive a precise continuous-time stochastic process to describe the system behaviors. To solve the above problem, we apply Variational Bayesian method and propose a flexible continuous-time stochastic recurrent neural network named Variational Stochastic Differential Networks (VSDN), which embeds the complicated dynamics of the sporadic time series by neural Stochastic Differential Equations (SDE). VSDNs capture the stochastic dependency among latent states and observations by deep neural networks. We also incorporate two differential Evidence Lower Bounds to efficiently train the models. Through comprehensive experiments, we show that VSDNs outperform state-of-the-art continuous-time deep learning models and achieve remarkable performance on prediction and interpolation tasks for sporadic time series.

cs.LG↗

arXiv2021

Training Noisy Single-Channel Speech Separation With Noisy Oracle Sources: A Large Gap and A Small Step

As the performance of single-channel speech separation systems has improved, there has been a desire to move to more challenging conditions than the clean, near-field speech that initial systems were developed on. When training deep learning separation models, a need for ground truth leads to training on synthetic mixtures. As such, training in noisy conditions requires either using noise synthetically added to clean speech, preventing the use of in-domain data for a noisy-condition task, or training using mixtures of noisy speech, requiring the network to additionally separate the noise. We demonstrate the relative inseparability of noise and that this noisy speech paradigm leads to significant degradation of system performance. We also propose an SI-SDR-inspired training objective that tries to exploit the inseparability of noise to implicitly partition the signal and discount noise separation errors, enabling the training of better separation systems with noisy oracle sources.

eess.AS↗