arXiv subjects

Xin Wang

Publications and source records attributed to Xin Wang.

At least 991 records · Page 55Linked to original sources

arXiv2024

UMedNeRF: Uncertainty-aware Single View Volumetric Rendering for Medical Neural Radiance Fields

In the field of clinical medicine, computed tomography (CT) is an effective medical imaging modality for the diagnosis of various pathologies. Compared with X-ray images, CT images can provide more information, including multi-planar slices and three-dimensional structures for clinical diagnosis. However, CT imaging requires patients to be exposed to large doses of ionizing radiation for a long time, which may cause irreversible physical harm. In this paper, we propose an Uncertainty-aware MedNeRF (UMedNeRF) network based on generated radiation fields. The network can learn a continuous representation of CT projections from 2D X-ray images by obtaining the internal structure and depth information and using adaptive loss weights to ensure the quality of the generated images. Our model is trained on publicly available knee and chest datasets, and we show the results of CT projection rendering with a single X-ray and compare our method with other methods based on generated radiation fields.

eess.IV↗

arXiv2024

A massive interacting galaxy 510 million years after the Big Bang

JWST observations confirm the existence of galaxies as early as 300Myr and at a higher number density than expected based on galaxy formation models and HST observations. Yet, sources confirmed spectroscopically in the first 500Myr have estimated stellar masses $<5\times10^8M_\odot$, limiting the signal to noise ratio (SNR) for investigating substructure. We present a high-resolution spectroscopic and spatially resolved study of a rare bright galaxy at $z=9.3127\pm0.0002$ with a stellar mass of $(2.5^{+0.7}_{-0.5})\times10^9M_\odot$, forming $25^{+3}_{-4}M_\odot/yr$ and with a metallicity of $\sim0.1Z_\odot$- lower than in the local universe for the stellar mass but in line with expectations of chemical enrichment in galaxies 1-2Gyr after the Big Bang. The system has a morphology typically associated to two interacting galaxies, with a two-component main clump of very young stars (age$<10$Myr) surrounded by an extended stellar population ($130\pm20$Myr old, identified by modeling the NIRSpec spectrum) and an elongated clumpy tidal tail. The spectroscopic observations identify O, Ne and H emission lines, and the Lyman break, where there is evidence of substantial Ly$α$ absorption. The [OII] doublet is resolved spectrally, enabling an estimate of the electron number density and ionization parameter of the interstellar medium and showing higher densities and ionization than in lower redshift analogs. For the first time at $z>8$, we identify evidence of absorption lines (Si, C and Fe), with low confidence individual detections but SNR$>6$ when stacked. The absorption features suggest that Ly$α$ is damped by the interstellar and circumgalactic medium. Our observations provide evidence of rapid efficient build-up of mass and metals in the immediate aftermath of the Big Bang through mergers, demonstrating that massive galaxies with several billion stars exist earlier than expected.

astro-ph.GA↗

arXiv2024

DisenBooth: Identity-Preserving Disentangled Tuning for Subject-Driven Text-to-Image Generation

Subject-driven text-to-image generation aims to generate customized images of the given subject based on the text descriptions, which has drawn increasing attention. Existing methods mainly resort to finetuning a pretrained generative model, where the identity-relevant information (e.g., the boy) and the identity-irrelevant information (e.g., the background or the pose of the boy) are entangled in the latent embedding space. However, the highly entangled latent embedding may lead to the failure of subject-driven text-to-image generation as follows: (i) the identity-irrelevant information hidden in the entangled embedding may dominate the generation process, resulting in the generated images heavily dependent on the irrelevant information while ignoring the given text descriptions; (ii) the identity-relevant information carried in the entangled embedding can not be appropriately preserved, resulting in identity change of the subject in the generated images. To tackle the problems, we propose DisenBooth, an identity-preserving disentangled tuning framework for subject-driven text-to-image generation. Specifically, DisenBooth finetunes the pretrained diffusion model in the denoising process. Different from previous works that utilize an entangled embedding to denoise each image, DisenBooth instead utilizes disentangled embeddings to respectively preserve the subject identity and capture the identity-irrelevant information. We further design the novel weak denoising and contrastive embedding auxiliary tuning objectives to achieve the disentanglement. Extensive experiments show that our proposed DisenBooth framework outperforms baseline models for subject-driven text-to-image generation with the identity-preserved embedding. Additionally, by combining the identity-preserved embedding and identity-irrelevant embedding, DisenBooth demonstrates more generation flexibility and controllability

cs.CV↗

arXiv2024

Preserving Fairness Generalization in Deepfake Detection

Although effective deepfake detection models have been developed in recent years, recent studies have revealed that these models can result in unfair performance disparities among demographic groups, such as race and gender. This can lead to particular groups facing unfair targeting or exclusion from detection, potentially allowing misclassified deepfakes to manipulate public opinion and undermine trust in the model. The existing method for addressing this problem is providing a fair loss function. It shows good fairness performance for intra-domain evaluation but does not maintain fairness for cross-domain testing. This highlights the significance of fairness generalization in the fight against deepfakes. In this work, we propose the first method to address the fairness generalization problem in deepfake detection by simultaneously considering features, loss, and optimization aspects. Our method employs disentanglement learning to extract demographic and domain-agnostic forgery features, fusing them to encourage fair learning across a flattened loss landscape. Extensive experiments on prominent deepfake datasets demonstrate our method's effectiveness, surpassing state-of-the-art approaches in preserving fairness during cross-domain deepfake detection. The code is available at https://github.com/Purdue-M2/Fairness-Generalization

cs.CV↗

arXiv2024

Two-stage Cytopathological Image Synthesis for Augmenting Cervical Abnormality Screening

Automatic thin-prep cytologic test (TCT) screening can assist pathologists in finding cervical abnormality towards accurate and efficient cervical cancer diagnosis. Current automatic TCT screening systems mostly involve abnormal cervical cell detection, which generally requires large-scale and diverse training data with high-quality annotations to achieve promising performance. Pathological image synthesis is naturally raised to minimize the efforts in data collection and annotation. However, it is challenging to generate realistic large-size cytopathological images while simultaneously synthesizing visually plausible appearances for small-size abnormal cervical cells. In this paper, we propose a two-stage image synthesis framework to create synthetic data for augmenting cervical abnormality screening. In the first Global Image Generation stage, a Normal Image Generator is designed to generate cytopathological images full of normal cervical cells. In the second Local Cell Editing stage, normal cells are randomly selected from the generated images and then are converted to different types of abnormal cells using the proposed Abnormal Cell Synthesizer. Both Normal Image Generator and Abnormal Cell Synthesizer are built upon Stable Diffusion, a pre-trained foundation model for image synthesis, via parameter-efficient fine-tuning methods for customizing cytopathological image contents and extending spatial layout controllability, respectively. Our experiments demonstrate the synthetic image quality, diversity, and controllability of the proposed synthesis framework, and validate its data augmentation effectiveness in enhancing the performance of abnormal cervical cell detection.

cs.CV↗

arXiv2024

Theory of polygonal phases self-assembled from T-shaped liquid crystalline molecules

Extensive experimental studies have shown that numerous ordered phases can be formed via the self-assembly of T-shaped liquid crystalline molecules (TLCMs) composed of a rigid backbone, two flexible end chains and a flexible side chain. However, a comprehensive understanding of the stability and formation mechanisms of these intricately nanostructured phases remains incomplete. Here we fill this gap by carrying out a theoretical study of the phase behaviour of TLCMs. Specifically, we construct phase diagrams of TLCMs by computing the free energy of different ordered phases of the system. Our results reveal that the number of polygonal edges increases as the length of side chain or interaction strength increases, consistent with experimental observations. The theoretical study not only reproduces the experimentally observed phases and phase transition sequences, but also systematically analyzes the stability mechanism of the polygonal phases.

cond-mat.soft↗

arXiv2024

Circuit QED with a Giant Atom Coupling to Left-handed Superlattice Metamaterials

Giant atoms, where the dipole approximation ceases to be valid, allow us to observe unconventional quantum optical phenomena arising from interference and time-delay effects. Most previous studies consider giant atoms coupling to conventional materials with right-handed dispersion. In this study, we first investigate the quantum dynamics of a giant atom interacting with left-handed superlattice metamaterials. Different from those right-handed counterparts, the left-handed superlattices exhibit an asymmetric band gap generated by anomalous dispersive bands and Bragg scattering bands. First, by assuming that the giant atom is in resonance with the continuous dispersive energy band, spontaneous emission will undergo periodic enhancement or suppression due to the interference effect. At the resonant position, there is a significant discrepancy in the spontaneous decay rates between the upper and lower bands, which arises from the differences in group velocity. Second, we explore the non-Markovian dynamics of the giant atom by considering the frequency of the emitter outside the energy band, where bound states will be induced by the interference between two coupling points. By employing both analytical and numerical methods, we demonstrate that the steady atomic population will be periodically modulated, driven by variations in the size of the giant atom. The presence of asymmetric band edges leads to diverse interference dynamics. Finally, we consider the case of two identical emitters coupling to the waveguide and find that the energy within the two emitters undergoes exchange through the mechanism Rabi oscillations.

quant-ph↗

arXiv2024

A Survey on Fairness in Large Language Models

Large Language Models (LLMs) have shown powerful performance and development prospects and are widely deployed in the real world. However, LLMs can capture social biases from unprocessed training data and propagate the biases to downstream tasks. Unfair LLM systems have undesirable social impacts and potential harms. In this paper, we provide a comprehensive review of related research on fairness in LLMs. Considering the influence of parameter magnitude and training paradigm on research strategy, we divide existing fairness research into oriented to medium-sized LLMs under pre-training and fine-tuning paradigms and oriented to large-sized LLMs under prompting paradigms. First, for medium-sized LLMs, we introduce evaluation metrics and debiasing methods from the perspectives of intrinsic bias and extrinsic bias, respectively. Then, for large-sized LLMs, we introduce recent fairness research, including fairness evaluation, reasons for bias, and debiasing methods. Finally, we discuss and provide insight on the challenges and future directions for the development of fairness in LLMs.

cs.CL↗

arXiv2024

PokeMQA: Programmable knowledge editing for Multi-hop Question Answering

Multi-hop question answering (MQA) is one of the challenging tasks to evaluate machine's comprehension and reasoning abilities, where large language models (LLMs) have widely achieved the human-comparable performance. Due to the dynamics of knowledge facts in real world, knowledge editing has been explored to update model with the up-to-date facts while avoiding expensive re-training or fine-tuning. Starting from the edited fact, the updated model needs to provide cascading changes in the chain of MQA. The previous art simply adopts a mix-up prompt to instruct LLMs conducting multiple reasoning tasks sequentially, including question decomposition, answer generation, and conflict checking via comparing with edited facts. However, the coupling of these functionally-diverse reasoning tasks inhibits LLMs' advantages in comprehending and answering questions while disturbing them with the unskilled task of conflict checking. We thus propose a framework, Programmable knowledge editing for Multi-hop Question Answering (PokeMQA), to decouple the jobs. Specifically, we prompt LLMs to decompose knowledge-augmented multi-hop question, while interacting with a detached trainable scope detector to modulate LLMs behavior depending on external conflict signal. The experiments on three LLM backbones and two benchmark datasets validate our superiority in knowledge editing of MQA, outperforming all competitors by a large margin in almost all settings and consistently producing reliable reasoning process.

cs.CL↗

arXiv2024

On-the-Fly SfM: What you capture is What you get

Over the last decades, ample achievements have been made on Structure from motion (SfM). However, the vast majority of them basically work in an offline manner, i.e., images are firstly captured and then fed together into a SfM pipeline for obtaining poses and sparse point cloud. In this work, on the contrary, we present an on-the-fly SfM: running online SfM while image capturing, the newly taken On-the-Fly image is online estimated with the corresponding pose and points, i.e., what you capture is what you get. Specifically, our approach firstly employs a vocabulary tree that is unsupervised trained using learning-based global features for fast image retrieval of newly fly-in image. Then, a robust feature matching mechanism with least squares (LSM) is presented to improve image registration performance. Finally, via investigating the influence of newly fly-in image's connected neighboring images, an efficient hierarchical weighted local bundle adjustment (BA) is used for optimization. Extensive experimental results demonstrate that on-the-fly SfM can meet the goal of robustly registering the images while capturing in an online way.

cs.CV↗

arXiv2024

Fast entangling quantum gates with almost-resonant modulated driving

Recently, the method of off-resonant modulated driving (ORMD) with a special category of synthetic analytical pulses has improved the experimental performance of two- and multi-qubit gates and aroused many interests for further investigations. It particularly offers a helpful tool to the cold atom qubit platform and works well with the Rydberg dipole-dipole interaction. In order to explore more possibilities and wider ranges of options in constructing fast-speed and high-fidelity quantum logic gates, we design and analyze the entangling quantum gates via the almost-resonant modulated driving (ARMD) method. Apart from the apparent distinctions in resonance conditions, the ARMD gate protocols also have its different mechanisms in quantum physics compared with ORMD gate protocols. ARMD gates usually have abrupt phase changes and at certain points during the time evolution. In other words, whilst the modulation forms the key concept of high-fidelity Rydberg blockade gates, the on-off resonance condition can lead to nontrivial nuances in the styles of dynamics. From a more fundamental point of view, the ORMD and the ARMD methods all together belong to the unitary operation family of fast modulated driving with respect to precisely characterized inter-qubit interactions, which usually allows the quantum logic gate to concludes within one continuous pulse.

quant-ph↗

arXiv2024

Follow-up on the Supermassive Black Hole Binary Candidate J1048+7143: Successful Prediction of the Next Gamma-ray Flare and Refined Binary Parameters in the Framework of Jet Precession Model

Analyzing single-dish and VLBI radio, as well as \textit{Fermi}-LAT $γ$-ray observations, we explained the three major flares in the $γ$-ray light curve of FSRQ J1048+7143 with the spin--orbit precession of the dominant mass black hole in a supermassive black hole binary system. Here, we report on the detection of a fourth $γ$-ray flare from J1048+7143, appearing in the time interval which was predicted in our previous work. Including this new flare, we constrained the mass ratio into a narrow range of $0.062<q<0.088$, and consequently we were able to further constrain the parameters of the hypothetical supermassive binary black hole at the heart of J1048+7143. We predict the occurrence of the fifth major $γ$-ray flare that would appear only if the jet will still lay close to our line sight. The fourth major $γ$-ray flare also shows the two-subflare structure, further strengthening our scenario in which the occurrence of the subflares is the signature of the precession of a spine--sheath jet structure that quasi-periodically interacts with a proton target, e.g. clouds in the broad-line region.

astro-ph.HE↗

arXiv2024

Rethinking Propagation for Unsupervised Graph Domain Adaptation

Unsupervised Graph Domain Adaptation (UGDA) aims to transfer knowledge from a labelled source graph to an unlabelled target graph in order to address the distribution shifts between graph domains. Previous works have primarily focused on aligning data from the source and target graph in the representation space learned by graph neural networks (GNNs). However, the inherent generalization capability of GNNs has been largely overlooked. Motivated by our empirical analysis, we reevaluate the role of GNNs in graph domain adaptation and uncover the pivotal role of the propagation process in GNNs for adapting to different graph domains. We provide a comprehensive theoretical analysis of UGDA and derive a generalization bound for multi-layer GNNs. By formulating GNN Lipschitz for k-layer GNNs, we show that the target risk bound can be tighter by removing propagation layers in source graph and stacking multiple propagation layers in target graph. Based on the empirical and theoretical analysis mentioned above, we propose a simple yet effective approach called A2GNN for graph domain adaptation. Through extensive experiments on real-world datasets, we demonstrate the effectiveness of our proposed A2GNN framework.

cs.LG↗

arXiv2024

Failure Analysis in Next-Generation Critical Cellular Communication Infrastructures

The advent of communication technologies marks a transformative phase in critical infrastructure construction, where the meticulous analysis of failures becomes paramount in achieving the fundamental objectives of continuity, security, and availability. This survey enriches the discourse on failures, failure analysis, and countermeasures in the context of the next-generation critical communication infrastructures. Through an exhaustive examination of existing literature, we discern and categorize prominent research orientations with focuses on, namely resource depletion, security vulnerabilities, and system availability concerns. We also analyze constructive countermeasures tailored to address identified failure scenarios and their prevention. Furthermore, the survey emphasizes the imperative for standardization in addressing failures related to Artificial Intelligence (AI) within the ambit of the sixth-generation (6G) networks, accounting for the forward-looking perspective for the envisioned intelligence of 6G network architecture. By identifying new challenges and delineating future research directions, this survey can help guide stakeholders toward unexplored territories, fostering innovation and resilience in critical communication infrastructure development and failure prevention.

eess.SY↗

arXiv2024

Revisiting VAE for Unsupervised Time Series Anomaly Detection: A Frequency Perspective

Time series Anomaly Detection (AD) plays a crucial role for web systems. Various web systems rely on time series data to monitor and identify anomalies in real time, as well as to initiate diagnosis and remediation procedures. Variational Autoencoders (VAEs) have gained popularity in recent decades due to their superior de-noising capabilities, which are useful for anomaly detection. However, our study reveals that VAE-based methods face challenges in capturing long-periodic heterogeneous patterns and detailed short-periodic trends simultaneously. To address these challenges, we propose Frequency-enhanced Conditional Variational Autoencoder (FCVAE), a novel unsupervised AD method for univariate time series. To ensure an accurate AD, FCVAE exploits an innovative approach to concurrently integrate both the global and local frequency features into the condition of Conditional Variational Autoencoder (CVAE) to significantly increase the accuracy of reconstructing the normal data. Together with a carefully designed "target attention" mechanism, our approach allows the model to pick the most useful information from the frequency domain for better short-periodic trend construction. Our FCVAE has been evaluated on public datasets and a large-scale cloud system, and the results demonstrate that it outperforms state-of-the-art methods. This confirms the practical applicability of our approach in addressing the limitations of current VAE-based anomaly detection models.

cs.LG↗

arXiv2024

Masked Conditional Diffusion Model for Enhancing Deepfake Detection

Recent studies on deepfake detection have achieved promising results when training and testing faces are from the same dataset. However, their results severely degrade when confronted with forged samples that the model has not yet seen during training. In this paper, deepfake data to help detect deepfakes. this paper present we put a new insight into diffusion model-based data augmentation, and propose a Masked Conditional Diffusion Model (MCDM) for enhancing deepfake detection. It generates a variety of forged faces from a masked pristine one, encouraging the deepfake detection model to learn generic and robust representations without overfitting to special artifacts. Extensive experiments demonstrate that forgery images generated with our method are of high quality and helpful to improve the performance of deepfake detection models.

cs.CV↗

arXiv2024

Power Characterization of Noisy Quantum Kernels

Quantum kernel methods have been widely recognized as one of promising quantum machine learning algorithms that have potential to achieve quantum advantages. In this paper, we theoretically characterize the power of noisy quantum kernels and demonstrate that under global depolarization noise, for different input data the predictions of the optimal hypothesis inferred by the noisy quantum kernel approximately concentrate towards some fixed value. In particular, we depict the convergence rate in terms of the strength of quantum noise, the size of training samples, the number of qubits, the number of layers affected by quantum noises, as well as the number of measurement shots. Our results show that noises may make quantum kernel methods to only have poor prediction capability, even when the generalization error is small. Thus, we provide a crucial warning to employ noisy quantum kernel methods for quantum computation and the theoretical results can also serve as guidelines when developing practical quantum kernel algorithms for achieving quantum advantages.

quant-ph↗

arXiv2024

Uncertainty-Aware Explainable Recommendation with Large Language Models

Providing explanations within the recommendation system would boost user satisfaction and foster trust, especially by elaborating on the reasons for selecting recommended items tailored to the user. The predominant approach in this domain revolves around generating text-based explanations, with a notable emphasis on applying large language models (LLMs). However, refining LLMs for explainable recommendations proves impractical due to time constraints and computing resource limitations. As an alternative, the current approach involves training the prompt rather than the LLM. In this study, we developed a model that utilizes the ID vectors of user and item inputs as prompts for GPT-2. We employed a joint training mechanism within a multi-task learning framework to optimize both the recommendation task and explanation task. This strategy enables a more effective exploration of users' interests, improving recommendation effectiveness and user satisfaction. Through the experiments, our method achieving 1.59 DIV, 0.57 USR and 0.41 FCR on the Yelp, TripAdvisor and Amazon dataset respectively, demonstrates superior performance over four SOTA methods in terms of explainability evaluation metric. In addition, we identified that the proposed model is able to ensure stable textual quality on the three public datasets.

cs.IR↗