arXiv subjects

Qian Wang

Publications and source records attributed to Qian Wang.

At least 379 records · Page 21Linked to original sources

arXiv2023

MDP: A Generalized Framework for Text-Guided Image Editing by Manipulating the Diffusion Path

Image generation using diffusion can be controlled in multiple ways. In this paper, we systematically analyze the equations of modern generative diffusion networks to propose a framework, called MDP, that explains the design space of suitable manipulations. We identify 5 different manipulations, including intermediate latent, conditional embedding, cross attention maps, guidance, and predicted noise. We analyze the corresponding parameters of these manipulations and the manipulation schedule. We show that some previous editing methods fit nicely into our framework. Particularly, we identified one specific configuration as a new type of control by manipulating the predicted noise, which can perform higher-quality edits than previous work for a variety of local and global edits.

cs.CV↗

arXiv2023

BlobGAN-3D: A Spatially-Disentangled 3D-Aware Generative Model for Indoor Scenes

3D-aware image synthesis has attracted increasing interest as it models the 3D nature of our real world. However, performing realistic object-level editing of the generated images in the multi-object scenario still remains a challenge. Recently, a 2D GAN termed BlobGAN has demonstrated great multi-object editing capabilities on real-world indoor scene datasets. In this work, we propose BlobGAN-3D, which is a 3D-aware improvement of the original 2D BlobGAN. We enable explicit camera pose control while maintaining the disentanglement for individual objects in the scene by extending the 2D blobs into 3D blobs. We keep the object-level editing capabilities of BlobGAN and in addition allow flexible control over the 3D location of the objects in the scene. We test our method on real-world indoor datasets and show that our method can achieve comparable image quality compared to the 2D BlobGAN and other 3D-aware GAN baselines while being able to enable camera pose control and object-level editing in the challenging multi-object real-world scenarios.

cs.CV↗

arXiv2023

Spatial Attention-based Implicit Neural Representation for Arbitrary Reduction of MRI Slice Spacing

Magnetic resonance (MR) images collected in 2D clinical protocols typically have large inter-slice spacing, resulting in high in-plane resolution and reduced through-plane resolution. Super-resolution technique can enhance the through-plane resolution of MR images to facilitate downstream visualization and computer-aided diagnosis. However, most existing works train the super-resolution network at a fixed scaling factor, which is not friendly to clinical scenes of varying inter-slice spacing in MR scanning. Inspired by the recent progress in implicit neural representation, we propose a Spatial Attention-based Implicit Neural Representation (SA-INR) network for arbitrary reduction of MR inter-slice spacing. The SA-INR aims to represent an MR image as a continuous implicit function of 3D coordinates. In this way, the SA-INR can reconstruct the MR image with arbitrary inter-slice spacing by continuously sampling the coordinates in 3D space. In particular, a local-aware spatial attention operation is introduced to model nearby voxels and their affinity more accurately in a larger receptive field. Meanwhile, to improve the computational efficiency, a gradient-guided gating mask is proposed for applying the local-aware spatial attention to selected areas only. We evaluate our method on the public HCP-1200 dataset and the clinical knee MR dataset to demonstrate its superiority over other existing methods.

eess.IV↗

arXiv2023

Excited state quantum phase transition and Loschmidt echo spectra in a spinor Bose-Einstein condensate

Identifying dynamical signatures of excited state quantum phase transitions (ESQPTs) in experimentally realizable quantum many-body systems is helpful for understanding the dynamical effects of ESQPTs. In such systems, the highly controllable spinor Bose-Einstein condensates (BECs) offer an exceptional platform to study ESQPTs. In this work, we investigate the dynamical characteristics of the ESQPT in spin-$1$ BEC by means of the Loschmidt echo spectrum. The Loschmidt echo spectrum is an extension of the well-known Loschmidt echo and definded as the overlaps between the evolved state and the excited states of the initial Hamiltonian. We show that both the time evolved and long time averaged Loschmidt echo spectrum undergo a remarkable change as the system passes through the critical point of the ESQPT. Moreover, the particular behavior exhibited by the Loschmidt echo spectrum at the critical point stand as a dynamical detector for probing the ESQPT. We further demonstrate how to capture the features of the ESQPT by using the energy distribution associated with the Loschmidt echo spectrum for time evolved and long time averaged cases, respectrively. Our findings contribute to a further verification of the usefulness of the Loschmidt echo spectrum for witnessing various quantum phase transitions in many-body systems and provide a new way to experimentally examine the dynamical consequences of ESQPTs.

cond-mat.quant-gas↗

arXiv2023

Deep learning reveals the common spectrum underlying multiple brain disorders in youth and elders from brain functional networks

Brain disorders in the early and late life of humans potentially share pathological alterations in brain functions. However, the key evidence from neuroimaging data for pathological commonness remains unrevealed. To explore this hypothesis, we build a deep learning model, using multi-site functional magnetic resonance imaging data (N=4,410, 6 sites), for classifying 5 different brain disorders from healthy controls, with a set of common features. Our model achieves 62.6(1.9)% overall classification accuracy on data from the 6 investigated sites and detects a set of commonly affected functional subnetworks at different spatial scales, including default mode, executive control, visual, and limbic networks. In the deep-layer feature representation for individual data, we observe young and aging patients with disorders are continuously distributed, which is in line with the clinical concept of the "spectrum of disorders". The revealed spectrum underlying early- and late-life brain disorders promotes the understanding of disorder comorbidities in the lifespan.

cs.AI↗

arXiv2023

An End-to-End Network for Co-Saliency Detection in One Single Image

Co-saliency detection within a single image is a common vision problem that has received little attention and has not yet been well addressed. Existing methods often used a bottom-up strategy to infer co-saliency in an image in which salient regions are firstly detected using visual primitives such as color and shape and then grouped and merged into a co-saliency map. However, co-saliency is intrinsically perceived complexly with bottom-up and top-down strategies combined in human vision. To address this problem, this study proposes a novel end-to-end trainable network comprising a backbone net and two branch nets. The backbone net uses ground-truth masks as top-down guidance for saliency prediction, whereas the two branch nets construct triplet proposals for regional feature mapping and clustering, which drives the network to be bottom-up sensitive to co-salient regions. We construct a new dataset of 2,019 natural images with co-saliency in each image to evaluate the proposed method. Experimental results show that the proposed method achieves state-of-the-art accuracy with a running speed of 28 fps.

cs.CV↗

arXiv2023

ChatCAD: Interactive Computer-Aided Diagnosis on Medical Image using Large Language Models

Large language models (LLMs) have recently demonstrated their potential in clinical applications, providing valuable medical knowledge and advice. For example, a large dialog LLM like ChatGPT has successfully passed part of the US medical licensing exam. However, LLMs currently have difficulty processing images, making it challenging to interpret information from medical images, which are rich in information that supports clinical decisions. On the other hand, computer-aided diagnosis (CAD) networks for medical images have seen significant success in the medical field by using advanced deep-learning algorithms to support clinical decision-making. This paper presents a method for integrating LLMs into medical-image CAD networks. The proposed framework uses LLMs to enhance the output of multiple CAD networks, such as diagnosis networks, lesion segmentation networks, and report generation networks, by summarizing and reorganizing the information presented in natural language text format. The goal is to merge the strengths of LLMs' medical domain knowledge and logical reasoning with the vision understanding capability of existing medical-image CAD models to create a more user-friendly and understandable system for patients compared to conventional CAD systems. In the future, LLM's medical knowledge can be also used to improve the performance of vision-based medical-image CAD models.

cs.CV↗

arXiv2023

The failure of semiclassical approach in the dissipative fully-connected Ising model

We solve the fully-connected Ising model in the presence of dissipation and time-periodic field, with the corresponding Lindblad equation having a time-periodic Liouvillian. The dynamics of the magnetizations is studied by using both the semiclassical approach and the numerical simulation with the help of permutation symmetry. The semiclassical approach shows a transition from the periodic response for small field amplitude to the chaotic dynamics for large amplitude. The trajectory of the magnetizations and the Lyapunov exponents are calculated, which support the existence of a chaotic phase. But in the exact numerical simulation, the response is periodic for both small and large amplitude. The scaling analysis of Floquet Liouvillian spectrum confirms the periodic response in the thermodynamic limit. The semiclassical approximation is found to fail as the field amplitude is large.

nlin.CD↗

arXiv2023

Simulation and Measurement of Out-of-Band Resonances for the FDM Readout of a TES Bolometer

With applications in cosmology, infrared astronomy and CMB survey, frequency-division multiplexing (FDM) proved to be a viable readout for transition-edge sensors (TES). We investigate the occurrence of out-of-band resonances (OBR) which could constrain the bandwidth of the FDM readout of TES bolometers. The study includes SPICE modeling of the entire setup including the cryogenic harness, LC filters, Superconducting Quantum Interference Device (SQUID) and room-temperature amplifier. Simulation results show that the long harness (for flight model) could cause multiple reflections that generate repetitive spikes in the spectrum. Peaks of the OBR are mainly due to the parasitic capacitances at the input of SQUID. Implementing a low-pass RC circuit (snubber) at the input of the SQUID dampened the OBR. As a result, the first peak only appears around 20 MHz which is a safe margin for the 1 MHz-3.8 MHz FDM in use in the prototype readout. Using a spectrum analyzer and broadband LNAs,we also measured the OBR for the prototype FDM readout in the lab up to 500 MHz. The measurement was conducted at temperatures of 50 mK and 4 K and for various biasing of the DC SQUID. It turns out that OBR are more intense at 50 mK and are caused by the harness impedance mismatch rather than the SQUID. Simulation codes and supporting materials are available at https://github.com/githubamin/LT-Spice-Simulation-of-FDM-readout.

astro-ph.IM↗

arXiv2022

Proposal for the search for exotic spin-spin interactions at the micrometer scale using functionalized cantilever force sensors

Spin-dependent exotic interactions can be generated by exchanging hypothetical bosons, which were introduced to solve some puzzles in physics. Many precision experiments have been performed to search for such interactions, but no confirmed observation has been made. Here, we propose new experiments to search for the exotic spin-spin interactions that can be mediated by axions or Z$^\prime$ bosons. A sensitive functionalized cantilever is utilized as a force sensor to measure the interactions between the spin-polarized electrons in a periodic magnetic source structure and a closed-loop magnetic structure integrated on the cantilever. The source is set to oscillate during data acquisition to modulate the exotic force signal to high harmonics of the oscillating frequency. This helps to suppress the spurious signals at the signal frequency. Different magnetic source structures are designed for different interaction detections. A magnetic stripe structure is designed for Z$^\prime$-mediated interaction, which is insensitive to the detection of axion-mediated interaction. This allows us to measure the coupling constant of both if we assume both exist. With the force sensitivity achievable at low temperature, the proposed experiments are expected to search for the parameter spaces with much smaller coupling constant than the current stringent constraints from micrometer to millimeter range. Specifically, the lower bound of the parameter space will be seven orders of magnitude lower than the stringent constraints for Z$^\prime$-mediated interaction, and an order of magnitude lower for axion-mediated interaction, at the interaction range of $10\, μ$m.

hep-ex↗

arXiv2022

Probabilistic Constraint Tightening Techniques for Trajectory Planning with Predictive Control

In order for automated mobile vehicles to navigate in the real world with minimal collision risks, it is necessary for their planning algorithms to consider uncertainties from measurements and environmental disturbances. In this paper, we consider analytical solutions for a conservative approximation of the mutual probability of collision between two robotic vehicles in the presence of such uncertainties. Therein, we present two methods, which we call unitary scaling and principal axes rotation, for decoupling the bivariate integral required for efficient approximation of the probability of collision between two vehicles including orientation effects. We compare the conservatism of these methods analytically and numerically. By closing a control loop through a model predictive guidance scheme, we observe through Monte-Carlo simulations that directly implementing collision avoidance constraints from the conservative approximations remains infeasible for real-time planning. We then propose and implement a convexification approach based on the tightened collision constraints that significantly improves the computational efficiency and robustness of the predictive guidance scheme.

cs.RO↗

arXiv2022

Integrating multi-type aberrations from DNA and RNA through dynamic mapping gene space for subtype-specific breast cancer driver discovery

Driver event discovery is a crucial demand for breast cancer diagnosis and therapy. Especially, discovering subtype-specificity of drivers can prompt the personalized biomarker discovery and precision treatment of cancer patients. still, most of the existing computational driver discovery studies mainly exploit the information from DNA aberrations and gene interactions. Notably, cancer driver events would occur due to not only DNA aberrations but also RNA alternations, but integrating multi-type aberrations from both DNA and RNA is still a challenging task for breast cancer drivers. On the one hand, the data formats of different aberration types also differ from each other, known as data format incompatibility. One the other hand, different types of aberrations demonstrate distinct patterns across samples, known as aberration type heterogeneity. To promote the integrated analysis of subtype-specific breast cancer drivers, we design a "splicing-and-fusing" framework to address the issues of data format incompatibility and aberration type heterogeneity respectively. To overcome the data format incompatibility, the "splicing-step" employs a knowledge graph structure to connect multi-type aberrations from the DNA and RNA data into a unified formation. To tackle the aberration type heterogeneity, the "fusing-step" adopts a dynamic mapping gene space integration approach to represent the multi-type information by vectorized profiles. The experiments also demonstrate the advantages of our approach in both the integration of multi-type aberrations from DNA and RNA and the discovery of subtype-specific breast cancer drivers. In summary, our "splicing-and-fusing" framework with knowledge graph connection and dynamic mapping gene space fusion of multi-type aberrations data from DNA and RNA can successfully discover potential breast cancer drivers with subtype-specificity indication.

q-bio.QM↗

arXiv2022

Nature of $X(3872)$ in $B^0 \to K^0 X(3872)$ and $B^+ \to K^+ X(3872)$ decays

We investigate the decays of $B^0 \to K^0 X(3872)$ and $B^+ \to K^+ X(3872)$ based on the picture where the $X(3872)$ resonance is strongly coupled to the $D\bar{D}^* + c.c.$ channel. In addition to the decay mechanism where the $X(3872)$ resonance is formed from the $c\bar{c}$ pair hadronization with the short-distance interaction, we have also considered the $D\bar{D}^*$ rescattering diagrams in the long-distance scale, where $D$ and $\bar{D}^*$ are formed from $c$ and $\bar{c}$ separately. Because of the difference of the mass thresholds of charged and neutral $D\bar{D}^*$ channels, and the rather narrow width of the $X(3872)$ resonance, at the $X(3872)$ mass, the loop functions of $D^0\bar{D}^{*0}$ and $D^+\bar{D}^{*-}$ are much different. Taking this difference into account, the ratio of $\mathcal{B}[B^0\to K^0X(3872)]/\mathcal{B}[B^+ \to K^+ X(3872)] \simeq 0.5$ can be naturally obtained. Based on this result, we also evaluate the decay widths of $B_s^0 \to η(η') X(3872)$. It is expected that future experimental measurements of these decays can be used to elucidate the nature of the $X(3872)$ resonance.

hep-ph↗

arXiv2022

On Fine-Tuned Deep Features for Unsupervised Domain Adaptation

Prior feature transformation based approaches to Unsupervised Domain Adaptation (UDA) employ the deep features extracted by pre-trained deep models without fine-tuning them on the specific source or target domain data for a particular domain adaptation task. In contrast, end-to-end learning based approaches optimise the pre-trained backbones and the customised adaptation modules simultaneously to learn domain-invariant features for UDA. In this work, we explore the potential of combining fine-tuned features and feature transformation based UDA methods for improved domain adaptation performance. Specifically, we integrate the prevalent progressive pseudo-labelling techniques into the fine-tuning framework to extract fine-tuned features which are subsequently used in a state-of-the-art feature transformation based domain adaptation method SPL (Selective Pseudo-Labeling). Thorough experiments with multiple deep models including ResNet-50/101 and DeiT-small/base are conducted to demonstrate the combination of fine-tuned features and SPL can achieve state-of-the-art performance on several benchmark datasets.

cs.CV↗

arXiv2022

Quantum Chaos in the Extended Dicke Model

We systematically study the chaotic signatures in a quantum many-body system consisting of an ensemble of interacting two-level atoms coupled to a single-mode bosonic field, the so-called extended Dicke model. The presence of the atom-atom interaction also leads us to explore how the atomic interaction affects the chaotic characters of the model. By analyzing the energy spectral statistics and the structure of eigenstates, we reveal the quantum signatures of chaos in the model and discuss the effect of the atomic interaction. We also investigate the dependence of the boundary of chaos extracted from both eigenvalue-based and eigenstate-based indicators on the atomic interaction. We show that the impact of the atomic interaction on the spectral statistics is stronger than on the structure of eigenstates. Qualitatively, the integrablity-to-chaos transition found in the Dicke model is amplified when the interatomic interaction in the extended Dicke model is switched on.

quant-ph↗

arXiv2022

Probabilistic Generative Transformer Language models for Generative Design of Molecules

Self-supervised neural language models have recently found wide applications in generative design of organic molecules and protein sequences as well as representation learning for downstream structure classification and functional prediction. However, most of the existing deep learning models for molecule design usually require a big dataset and have a black-box architecture, which makes it difficult to interpret their design logic. Here we propose Generative Molecular Transformer (GMTransformer), a probabilistic neural network model for generative design of molecules. Our model is built on the blank filling language model originally developed for text processing, which has demonstrated unique advantages in learning the "molecules grammars" with high-quality generation, interpretability, and data efficiency. Benchmarked on the MOSES datasets, our models achieve high novelty and Scaf compared to other baselines. The probabilistic generation steps have the potential in tinkering molecule design due to their capability of recommending how to modify existing molecules with explanation, guided by the learned implicit molecule chemistry. The source code and datasets can be accessed freely at https://github.com/usccolumbia/GMTransformer

cond-mat.mtrl-sci↗

arXiv2022

Determine the quantum numbers of $X(6900)$ from photon-photon fusion in ultra-peripheral heavy ion collisions

We study the production of the $X(6900)$ through the $J/ψJ/ψ$ decay channel in the ultra-peripheral heavy ion collisions at the LHC energy region. The potential quantum numbers of $X(6900)$ would be $0^{\pm +}$ and $2^{\pm+}$. We find that the transverse momentum and the polar angle distributions of $X(6900)$ can be used to distinguish these four potential quantum numbers. These typical distributions stem from the linearly polarized photon from the fast-moving nuclei and can be measured with the LHC experiments in further.

hep-ph↗

arXiv2022

Shuffled total least squares

Linear regression with shuffled labels and with a noisy latent design matrix arises in many correspondence recovery problems. We propose a total least-squares approach to the problem of estimating the underlying true permutation and provide an upper bound to the normalized Procrustes quadratic loss of the estimator. We also provide an iterative algorithm to approximate the estimator and demonstrate its performance on simulated data.

math.ST↗