arXiv subjects

Xin Wang

Publications and source records attributed to Xin Wang.

At least 1,027 records · Page 57Linked to original sources

arXiv2023

The Emergence of Brightest Cluster Galaxy in a Protocluster Core at $z=2.24$

We report the detection of a pair of massive quiescent galaxies likely in the process of merging at the center of the spectroscopically confirmed, extremely massive protocluster BOSS1244 at $z=2.24\pm0.02$. These galaxies, BOSS1244-QG1 and BOSS1244-QG2, were detected with Hubble Space Telescope (HST) grism slitless spectroscopic observations. These two quiescent galaxies are among the brightest member galaxies with $z=2.223-2.255$ in BOSS1244 and reside at redshifts $z=2.244$ and $z=2.242$, with a half-light radius of $6.76\pm0.50$ and $2.72\pm0.16$ kpc, respectively. BOSS1244-QG1 and BOSS1244-QG2 are separated by a projected distance of about 70 physical kpc, implying that the two galaxies likely merge to form a massive brightest cluster galaxy (BCG) with size and mass similar to the most massive BCGs in the local Universe. We thus infer that BCG formation through dry major mergers may happen earlier than the full assembly of a cluster core, which broadens our previous understanding of the co-evolution of mature galaxy clusters and BCGs in the nearby Universe. Moreover, we find a strong density-star formation relation over a scale of $\sim18$ co-moving Mpc in BOSS1244, i.e. star formation activity decreases as density increases, implying that the quenching of star formation in BCGs and their progenitors is likely governed by environment-related processes before the virialization of the cluster core.

astro-ph.GA↗

arXiv2023

JaxPruner: A concise library for sparsity research

This paper introduces JaxPruner, an open-source JAX-based pruning and sparse training library for machine learning research. JaxPruner aims to accelerate research on sparse neural networks by providing concise implementations of popular pruning and sparse training algorithms with minimal memory and latency overhead. Algorithms implemented in JaxPruner use a common API and work seamlessly with the popular optimization library Optax, which, in turn, enables easy integration with existing JAX based libraries. We demonstrate this ease of integration by providing examples in four different codebases: Scenic, t5x, Dopamine and FedJAX and provide baseline experiments on popular benchmarks.

cs.LG↗

arXiv2023

Nanospheres with Patches Arranged in Polyhedrons from Self-Assembly of Solution-State Diblock Copolymers under Spherical Confinement

Self-assembly of sphere-forming solution-state amphiphilic diblock copolymers under spherical nanopore confinement is investigated using a simulated annealing technique. For two types of cases of different pore-surface/copolymer interactions, sequences of self-assembled patchy nanospheres are obtained, and phase diagrams are constructed. Self-assembled patchy nanospheres with 1-21 solvophobic domains are observed. The outermost solvophobic domains (patches) are packed into various polyhedrons when their number is larger than 3, where three Platonic solids of a regular tetrahedron, an octahedron, and an icosahedron and seven Johnson solids of J12, J13, J17, J50, J51, J86, and J87 are identified. In addition, another Johnson solid of J84 is identified in a structure with two categories of B-domains. These polyhedrons have all or most of their faces in a triangular shape, and hence, they are closer to spherical in shape, which may relieve the chain stretching. Nanospheres with 1, 4, 6, 9, and 12 numbers of patches occur in relatively large windows in the phase diagrams of both types of cases. In one of the two types of cases, all nanospheres with any number of 1-14 patches occur in the phase diagram, whereas in the other type of cases, nanospheres with 2, 3, 5, 11, and 13 numbers of patches are absent in the phase diagram. Furthermore, at a given pore size, the number of patches changes nonmonotonically or is unchanged with an increase in the strength of the pore-surface/copolymer interactions for one type or the other type of case, respectively. Quantitative calculations are performed to elucidate mechanisms of the window size in the phase diagrams of nanospheres with different numbers of patches and structure details.

cond-mat.soft↗

arXiv2023

Towards Generalized Multi-stage Clustering: Multi-view Self-distillation

Existing multi-stage clustering methods independently learn the salient features from multiple views and then perform the clustering task. Particularly, multi-view clustering (MVC) has attracted a lot of attention in multi-view or multi-modal scenarios. MVC aims at exploring common semantics and pseudo-labels from multiple views and clustering in a self-supervised manner. However, limited by noisy data and inadequate feature learning, such a clustering paradigm generates overconfident pseudo-labels that mis-guide the model to produce inaccurate predictions. Therefore, it is desirable to have a method that can correct this pseudo-label mistraction in multi-stage clustering to avoid the bias accumulation. To alleviate the effect of overconfident pseudo-labels and improve the generalization ability of the model, this paper proposes a novel multi-stage deep MVC framework where multi-view self-distillation (DistilMVC) is introduced to distill dark knowledge of label distribution. Specifically, in the feature subspace at different hierarchies, we explore the common semantics of multiple views through contrastive learning and obtain pseudo-labels by maximizing the mutual information between views. Additionally, a teacher network is responsible for distilling pseudo-labels into dark knowledge, supervising the student network and improving its predictive capabilities to enhance the robustness. Extensive experiments on real-world multi-view datasets show that our method has better clustering performance than state-of-the-art methods.

cs.CV↗

arXiv2023

A Clustering Framework for Unsupervised and Semi-supervised New Intent Discovery

New intent discovery is of great value to natural language processing, allowing for a better understanding of user needs and providing friendly services. However, most existing methods struggle to capture the complicated semantics of discrete text representations when limited or no prior knowledge of labeled data is available. To tackle this problem, we propose a novel clustering framework, USNID, for unsupervised and semi-supervised new intent discovery, which has three key technologies. First, it fully utilizes unsupervised or semi-supervised data to mine shallow semantic similarity relations and provide well-initialized representations for clustering. Second, it designs a centroid-guided clustering mechanism to address the issue of cluster allocation inconsistency and provide high-quality self-supervised targets for representation learning. Third, it captures high-level semantics in unsupervised or semi-supervised data to discover fine-grained intent-wise clusters by optimizing both cluster-level and instance-level objectives. We also propose an effective method for estimating the cluster number in open-world scenarios without knowing the number of new intents beforehand. USNID performs exceptionally well on several benchmark intent datasets, achieving new state-of-the-art results in unsupervised and semi-supervised new intent discovery and demonstrating robust performance with different cluster numbers.

cs.CL↗

arXiv2023

Speaker-Text Retrieval via Contrastive Learning

In this study, we introduce a novel cross-modal retrieval task involving speaker descriptions and their corresponding audio samples. Utilizing pre-trained speaker and text encoders, we present a simple learning framework based on contrastive learning. Additionally, we explore the impact of incorporating speaker labels into the training process. Our findings establish the effectiveness of linking speaker and text information for the task for both English and Japanese languages, across diverse data configurations. Additional visual analysis unveils potential nuanced associations between speaker clustering and retrieval performance.

cs.SD↗

arXiv2023

The Ly$α$ non-detection by JWST NIRSpec of a strong Ly$α$ emitter at $z=5.66$ confirmed by MUSE

The detections of Lyman-$α$ ($\rm Lyα$) emission in galaxies with redshifts above 5 are of utmost importance for constraining the cosmic reionization timeline, yet such detections are usually based on slit spectroscopy. Here we investigate the significant bias induced by slit placement on the estimate of $\rm Lyα$ escape fraction ( $f_{\rm esc}^{\mathrm{Lyα}}$), by presenting a galaxy (dubbed A2744-z6Lya) at $z=5.66$ where its deep JWST NIRSpec prism spectroscopy completely misses the strong $\rm Lyα$ emission detected in the MUSE data. A2744-z6Lya exhibits a pronounced UV continuum with an extremely steep spectral slope of $β=-2.574_{-0.008}^{+0.008}$, and it has a stellar mass of $\mathrm{\sim10^{8.82}~M_\odot}$, a star-formation rate of $\mathrm{\sim8.35~M_\odot yr^{-1}}$ and gas-phase metallicity of $\mathrm{12+log\,(O/H)\sim7.88}$. The observed flux and rest-frame equivalent width of its Ly$α$ from MUSE spectroscopy are $1.2\times \rm 10^{-16} erg~s^{-1}cm^{-2}$ and 75Å, equivalent to $f_{\rm esc}^{\mathrm{Lyα}}=78\pm4 \%$. However, its Ly$α$ non-detection from JWST NIRSpec gives a 5-$σ$ upper limit of $<13 \%$, in stark contrast to that derived from MUSE. To explore the reasons for this bias, we perform spatially resolved stellar population analysis of A2744-z6Lya using the JWST NIRCam imaging data to construct 2-dimensional maps of SFR, dust extinction and neutral hydrogen column density. We find that the absence of Ly$α$ in the slit regions probably stems from both the resonance scattering effect of neutral hydrogen and dust extinction. Through analyzing an extreme case in detail, this work highlights the important caveat of inferring $f_{\rm esc}^{\mathrm{Lyα}}$ from slit spectroscopy, particularly when using the JWST multiplexed NIRSpec microshutter assembly.

astro-ph.GA↗

arXiv2023

Reversible Entanglement Beyond Quantum Operations

We introduce a reversible theory of exact entanglement manipulation by establishing a necessary and sufficient condition for state transfer under trace-preserving transformations that completely preserve the positivity of partial transpose (PPT). Under these free transformations, we show that logarithmic negativity emerges as the pivotal entanglement measure for determining entangled states' transformations, analogous to the role of entropy in the second law of thermodynamics. Previous results have proven that entanglement is irreversible under quantum operations that completely preserve PPT and leave open the question of reversibility for quantum operations that do not generate entanglement asymptotically. However, we find that going beyond the complete positivity constraint imposed by standard quantum mechanics enables a reversible theory of exact entanglement manipulation, which may suggest a potential incompatibility between the reversibility of entanglement and the fundamental principles of quantum mechanics.

quant-ph↗

arXiv2023

Axion and Dark Fermion Electromagnetic Form Factors in Superfluid He-4

Condensed matter materials have shown great potential in searching for light dark matter (DM) via detecting the phonon or magnon signals induced by the scattering of DMs off the materials. In this paper, we study the possibility of detecting electromagnetic form factors of fermionic DM and axion-like particles (ALPs) using superfluid Helium-4. The phonon induced by a sub-GeV fermionic DM scattering off the superfluid can be described using the effective field theory with the interaction between DM and the bulk ${}^4$He. Signals arising from the electromagnetic form factors of light DM in the presence of an external electric field are calculated. Projected constraints on the charge radius, the anapole moment, and the magnetic moment of the DM are derived with 1 kg$\cdot$year exposure. The phonon signal induced by the scattering of ALPs off the superfluid is also calculated, which can put competitive and the first direct detection bounds on ALP-photon-dark photon couplings in the projected experiments.

hep-ph↗

arXiv2023

Detection and Mitigation of Position Spoofing Attacks on Cooperative UAV Swarm Formations

Detecting spoofing attacks on the positions of unmanned aerial vehicles (UAVs) within a swarm is challenging. Traditional methods relying solely on individually reported positions and pairwise distance measurements are ineffective in identifying the misbehavior of malicious UAVs. This paper presents a novel systematic structure designed to detect and mitigate spoofing attacks in UAV swarms. We formulate the problem of detecting malicious UAVs as a localization feasibility problem, leveraging the reported positions and distance measurements. To address this problem, we develop a semidefinite relaxation (SDR) approach, which reformulates the non-convex localization problem into a convex and tractable semidefinite program (SDP). Additionally, we propose two innovative algorithms that leverage the proximity of neighboring UAVs to identify malicious UAVs effectively. Simulations demonstrate the superior performance of our proposed approaches compared to existing benchmarks. Our methods exhibit robustness across various swarm networks, showcasing their effectiveness in detecting and mitigating spoofing attacks. {\blue Specifically, the detection success rate is improved by up to 65\%, 55\%, and 51\% against distributed, collusion, and mixed attacks, respectively, compared to the benchmarks.

eess.SY↗

arXiv2023

Early results from GLASS-JWST. XXVII. The mass-metallicity relation in lensed field galaxies at cosmic noon with NIRISS

We present a measurement of the mass-metallicity relation (MZR) at cosmic noon, using the JWST near-infrared wide-field slitless spectroscopy obtained by the GLASS-JWST Early Release Science program. By combining the power of JWST and the lensing magnification by the foreground cluster A2744, we extend the measurements of the MZR to the dwarf mass regime at high redshifts. A sample of 50 galaxies with several emission lines is identified across two wide redshift ranges of $z=1.8-2.3$ and $2.6-3.4$ in the stellar mass range of $\log{(M_*/M_\odot)}\in [6.9, 10.0]$. The observed slope of MZR is $0.223 \pm 0.017$ and $0.294 \pm 0.010$ at these two redshift ranges, respectively, consistent with the slopes measured in field galaxies with higher masses. In addition, we assess the impact of the morphological broadening on emission line measurement by comparing two methods of using 2D forward modeling and line profile fitting to 1D extracted spectra. We show that ignoring the morphological broadening effect when deriving line fluxes from grism spectra results in a systematic reduction of flux by $\sim30\%$ on average. This discrepancy appears to affect all the lines and thus does not lead to significant changes in flux ratio and metallicity measurements. This assessment of the morphological broadening effect using JWST data presents, for the first time, an important guideline for future work deriving galaxy line fluxes from wide-field slitless spectroscopy, such as Euclid, Roman, and the Chinese Space Station Telescope.

astro-ph.GA↗

arXiv2023

A Unified Approach to Interpreting and Boosting Adversarial Transferability

In this paper, we use the interaction inside adversarial perturbations to explain and boost the adversarial transferability. We discover and prove the negative correlation between the adversarial transferability and the interaction inside adversarial perturbations. The negative correlation is further verified through different DNNs with various inputs. Moreover, this negative correlation can be regarded as a unified perspective to understand current transferability-boosting methods. To this end, we prove that some classic methods of enhancing the transferability essentially decease interactions inside adversarial perturbations. Based on this, we propose to directly penalize interactions during the attacking process, which significantly improves the adversarial transferability.

cs.LG↗

arXiv2023

Dissecting the Runtime Performance of the Training, Fine-tuning, and Inference of Large Language Models

Large Language Models (LLMs) have seen great advance in both academia and industry, and their popularity results in numerous open-source frameworks and techniques in accelerating LLM pre-training, fine-tuning, and inference. Training and deploying LLMs are expensive as it requires considerable computing resources and memory, hence many efficient approaches have been developed for improving system pipelines as well as operators. However, the runtime performance can vary significantly across hardware and software stacks, which makes it difficult to choose the best configuration. In this work, we aim to benchmark the performance from both macro and micro perspectives. First, we benchmark the end-to-end performance of pre-training, fine-tuning, and serving LLMs in different sizes , i.e., 7, 13, and 70 billion parameters (7B, 13B, and 70B) on three 8-GPU platforms with and without individual optimization techniques, including ZeRO, quantization, recomputation, FlashAttention. Then, we dive deeper to provide a detailed runtime analysis of the sub-modules, including computing and communication operators in LLMs. For end users, our benchmark and findings help better understand different optimization techniques, training and inference frameworks, together with hardware platforms in choosing configurations for deploying LLMs. For researchers, our in-depth module-wise analyses discover potential opportunities for future work to further optimize the runtime performance of LLMs.

cs.PF↗

arXiv2023

VTimeLLM: Empower LLM to Grasp Video Moments

Large language models (LLMs) have shown remarkable text understanding capabilities, which have been extended as Video LLMs to handle video data for comprehending visual details. However, existing Video LLMs can only provide a coarse description of the entire video, failing to capture the precise start and end time boundary of specific events. In this paper, we solve this issue via proposing VTimeLLM, a novel Video LLM designed for fine-grained video moment understanding and reasoning with respect to time boundary. Specifically, our VTimeLLM adopts a boundary-aware three-stage training strategy, which respectively utilizes image-text pairs for feature alignment, multiple-event videos to increase temporal-boundary awareness, and high-quality video-instruction tuning to further improve temporal understanding ability as well as align with human intents. Extensive experiments demonstrate that in fine-grained time-related comprehension tasks for videos such as Temporal Video Grounding and Dense Video Captioning, VTimeLLM significantly outperforms existing Video LLMs. Besides, benefits from the fine-grained temporal understanding of the videos further enable VTimeLLM to beat existing Video LLMs in video dialogue benchmark, showing its superior cross-modal understanding and reasoning abilities.

cs.CV↗

arXiv2023

Quantum Phase Processing and its Applications in Estimating Phase and Entropies

Quantum computing can provide speedups in solving many problems as the evolution of a quantum system is described by a unitary operator in an exponentially large Hilbert space. Such unitary operators change the phase of their eigenstates and make quantum algorithms fundamentally different from their classical counterparts. Based on this unique principle of quantum computing, we develop a new algorithmic toolbox "quantum phase processing" that can directly apply arbitrary trigonometric transformations to eigenphases of a unitary operator. The quantum phase processing circuit is constructed simply, consisting of single-qubit rotations and controlled-unitaries, typically using only one ancilla qubit. Besides the capability of phase transformation, quantum phase processing in particular can extract the eigen-information of quantum systems by simply measuring the ancilla qubit, making it naturally compatible with indirect measurement. Quantum phase processing complements another powerful framework known as quantum singular value transformation and leads to more intuitive and efficient quantum algorithms for solving problems that are particularly phase-related. As a notable application, we propose a new quantum phase estimation algorithm without quantum Fourier transform, which requires the fewest ancilla qubits and matches the best performance so far. We further exploit the power of our method by investigating a plethora of applications in Hamiltonian simulation, entanglement spectroscopy and quantum entropies estimation, demonstrating improvements or optimality for almost all cases.

quant-ph↗

arXiv2023

A Generic Stochastic Hybrid Car-following Model Based on Approximate Bayesian Computation

Car following (CF) models are fundamental to describing traffic dynamics. However, the CF behavior of human drivers is highly stochastic and nonlinear. As a result, identifying the best CF model has been challenging and controversial despite decades of research. Introduction of automated vehicles has further complicated this matter as their CF controllers remain proprietary, though their behavior appears different than human drivers. This paper develops a stochastic learning approach to integrate multiple CF models, rather than relying on a single model. The framework is based on approximate Bayesian computation that probabilistically concatenates a pool of CF models based on their relative likelihood of describing observed behavior. The approach, while data-driven, retains physical tractability and interpretability. Evaluation results using two datasets show that the proposed approach can better reproduce vehicle trajectories for both human driven and automated vehicles than any single CF model considered.

cs.LG↗

arXiv2023

OFDMA-F$^2$L: Federated Learning With Flexible Aggregation Over an OFDMA Air Interface

Federated learning (FL) can suffer from a communication bottleneck when deployed in mobile networks, limiting participating clients and deterring FL convergence. The impact of practical air interfaces with discrete modulations on FL has not previously been studied in depth. This paper proposes a new paradigm of flexible aggregation-based FL (F$^2$L) over orthogonal frequency division multiple-access (OFDMA) air interface, termed as ``OFDMA-F$^2$L'', allowing selected clients to train local models for various numbers of iterations before uploading the models in each aggregation round. We optimize the selections of clients, subchannels and modulations, adapting to channel conditions and computing powers. Specifically, we derive an upper bound on the optimality gap of OFDMA-F$^2$L capturing the impact of the selections, and show that the upper bound is minimized by maximizing the weighted sum rate of the clients per aggregation round. A Lagrange-dual based method is developed to solve this challenging mixed integer program of weighted sum rate maximization, revealing that a ``winner-takes-all'' policy provides the almost surely optimal client, subchannel, and modulation selections. Experiments on multilayer perceptrons and convolutional neural networks show that OFDMA-F$^2$L with optimal selections can significantly improve the training convergence and accuracy, e.g., by about 18\% and 5\%, compared to potential alternatives.

eess.SP↗

arXiv2023

Self-organized biodiversity in biotic resource systems

What determines biodiversity in nature is a prominent issue in ecology, especially in biotic resource systems that are typically devoid of cross-feeding. Here, we show that by incorporating pairwise encounters among consumer individuals within the same species, a multitude of consumer species can self-organize to coexist in a well-mixed system with one or a few biotic resource species. The coexistence modes can manifest as either stable steady states or self-organized oscillations. Importantly, all coexistence states are robust to stochasticity, whether employing the stochastic simulation algorithm or individual-based modeling. Our model quantitatively illustrates species distribution patterns across a wide range of ecological communities and can be broadly used to explain biodiversity in many biotic resource systems.

q-bio.PE↗