Searcharxiv⌕ Search

arXiv subjects

Bin Chen

Publications and source records attributed to Bin Chen.

At least 397 records · Page 22Linked to original sources
arXiv2023

Enhancing Keyphrase Generation by BART Finetuning with Splitting and Shuffling

Keyphrase generation is a task of identifying a set of phrases that best repre-sent the main topics or themes of a given text. Keyphrases are dividend int pre-sent and absent keyphrases. Recent approaches utilizing sequence-to-sequence models show effectiveness on absent keyphrase generation. However, the per-formance is still limited due to the hardness of finding absent keyphrases. In this paper, we propose Keyphrase-Focused BART, which exploits the differ-ences between present and absent keyphrase generations, and performs fine-tuning of two separate BART models for present and absent keyphrases. We further show effective approaches of shuffling keyphrases and candidate keyphrase ranking. For absent keyphrases, our Keyphrase-Focused BART achieved new state-of-the-art score on F1@5 in two out of five keyphrase gen-eration benchmark datasets.

cs.CL↗
arXiv2023

Holographic Complexity of Rotating Quantum Black Holes

We study holographic complexity for the rotating quantum BTZ black holes (quBTZ), the BTZ black holes with corrections from bulk quantum fields. Using double holography, the combined system of backreacted rotating BTZ black holes with conformal matters, can be holographically described by the rotating AdS4 C-metric with the BTZ black hole living on a codimension-1 brane. We investigate both volume complexity and action complexity of rotating quBTZ, and pay special attention to their late-time behaviors. When the mass of BTZ black hole is not very small and the rotation is not very slow, we show that the late-time rates of the volume complexity and the action complexity agree with each other up to a factor 2 and reduce to the ones of BTZ at the leading classical order, and they both receive subleading quantum corrections. For the volume complexity, the leading quantum correction comes from the backreaction of comformal matter on the geometry, similar to the static quBTZ case. For the action complexity, unlike the static case, the Wheeler-de Witt (WdW) patch in computing the action complexity for the rotating black hole does not touch the black hole singularity such that the leading order result is in good match with the one of classical BTZ. However, when the mass of BTZ black hole is small or the rotation parameter a is small, the quantum correction to the action complexity could be significant such that the late-time slope of the action complexity of quBTZ deviates very much from the one of classical BTZ. Remarkably, we notice that the nonrotating limit $a \to 0$ is singular and does not lead to the late-time slope of the action complexity for non-rotating quantum BTZ black hole. The similar phenomenon happens for higher dimensional rotating black holes.

hep-th↗
arXiv2023

GIFD: A Generative Gradient Inversion Method with Feature Domain Optimization

Federated Learning (FL) has recently emerged as a promising distributed machine learning framework to preserve clients' privacy, by allowing multiple clients to upload the gradients calculated from their local data to a central server. Recent studies find that the exchanged gradients also take the risk of privacy leakage, e.g., an attacker can invert the shared gradients and recover sensitive data against an FL system by leveraging pre-trained generative adversarial networks (GAN) as prior knowledge. However, performing gradient inversion attacks in the latent space of the GAN model limits their expression ability and generalizability. To tackle these challenges, we propose \textbf{G}radient \textbf{I}nversion over \textbf{F}eature \textbf{D}omains (GIFD), which disassembles the GAN model and searches the feature domains of the intermediate layers. Instead of optimizing only over the initial latent code, we progressively change the optimized layer, from the initial latent space to intermediate layers closer to the output images. In addition, we design a regularizer to avoid unreal image generation by adding a small ${l_1}$ ball constraint to the searching range. We also extend GIFD to the out-of-distribution (OOD) setting, which weakens the assumption that the training sets of GANs and FL tasks obey the same data distribution. Extensive experiments demonstrate that our method can achieve pixel-level reconstruction and is superior to the existing methods. Notably, GIFD also shows great generalizability under different defense strategy settings and batch sizes.

cs.CV↗
arXiv2023

Constructing Carrollian Field Theories from Null Reduction

In this paper, we propose a novel way to construct off-shell actions of $d$-dimensional Carrollian field theories by considering the null-reduction of the Bargmann invariant actions in $d+1$ dimensions. This is based on the fact that $d$-dimensional Carrollian symmetry is the restriction of the $(d+1)$-dimensional Bargmann symmetry to a null hyper-surface. We focus on free scalar field theory and electromagnetic field theory, and show that the electric and magnetic sectors of these theories originate from different Bargmann invariant actions in one higher dimension. In the cases of the massless free scalar field and $d=4$ electromagnetic field, we verify Carrollian conformal invariance of the resulting theories, and find that there appear naturally chain representations and staggered modules of Carrollian conformal algebra.

hep-th↗
arXiv2023

Thermodynamic Origins of Structural Metastability in Two-Dimensional Black Arsenic

Two-dimensional (2D) materials have aroused considerable research interests owing to their potential applications in nanoelectronics and optoelectronics. Thermodynamic stability of 2D structures inevitably affects the performance and power consumption of the fabricated nanodevices. Black arsenic (b-As), as a cousin of black phosphorus, has presented the extremely high anisotropy in physical properties. However, the systematic research on structural stability of b-As is still lack. Herein, we demonstrated the detailed analysis on structural metastability of the natural b-As, and determined its existence conditions in terms of two essential thermodynamic variables as hydrostatic pressure and temperature. Our results confirmed that b-As can only survive below 0.7 GPa, and then irreversibly transform to gray arsenic, in consistent with our theoretical calculations. Furthermore, thermal annealing strategy was developed to precisely control the thickness of b-As flake, and it sublimates at 300 oC. These results could pave the way for 2D b-As in many promising applications.

cond-mat.mtrl-sci↗
arXiv2023

Adjacency-hopping de Bruijn Sequences for Non-repetitive Coding

A special type of cyclic sequences named adjacency-hopping de Bruijn sequences is introduced in this paper. It is theoretically proved the existence of such sequences, and the number of such sequences is derived. These sequences guarantee that all neighboring codes are different while retaining the uniqueness of subsequences, which is a significant characteristic of original de Bruijn sequences in coding and matching. At last, the adjacency-hopping de Bruijn sequences are applied to structured light coding, and a color fringe pattern coded by such a sequence is presented. In summary, the proposed sequences demonstrate significant advantages in structured light coding by virtue of the uniqueness of subsequences and the adjacency-hopping characteristic, and show potential for extension to other fields with similar requirements of non-repetitive coding and efficient matching.

cs.IT↗
arXiv2023

The Minkowski problem in Heisenberg groups

As we all know, the Minkowski type problem is the cornerstone of the Brunn-Minkowski theory in Euclidean space. The Heisenberg group as a sub-Riemannian space is the simplest non-Abelian degenerate Riemannian space that is completely different from a Euclidean space. By analogy with the Minkowski type problem in Euclidean space, the Minkowski type problem in Heisenberg groups is still open. In the present paper, we develop for the first time a sub-Riemannian version of Minkowski type problem in the horizontal distributions of Heisenberg groups, and further give a positive answer to this sub-Riemannian Minkowski type problem via the variational method.

math.DG↗
arXiv2023

Backdoor Attack on Hash-based Image Retrieval via Clean-label Data Poisoning

A backdoored deep hashing model is expected to behave normally on original query images and return the images with the target label when a specific trigger pattern presents. To this end, we propose the confusing perturbations-induced backdoor attack (CIBA). It injects a small number of poisoned images with the correct label into the training data, which makes the attack hard to be detected. To craft the poisoned images, we first propose the confusing perturbations to disturb the hashing code learning. As such, the hashing model can learn more about the trigger. The confusing perturbations are imperceptible and generated by optimizing the intra-class dispersion and inter-class shift in the Hamming space. We then employ the targeted adversarial patch as the backdoor trigger to improve the attack performance. We have conducted extensive experiments to verify the effectiveness of our proposed CIBA. Our code is available at https://github.com/KuofengGao/CIBA.

cs.CV↗
arXiv2023

Path-integral quantization of tensionless (super) string

In this work, we study the tensionless (super)string in the formalism of path-integral quantization. We introduce BMS $bc$ and $βγ$ ghosts intrinsically by accounting for the Faddeev-Popov determinants appeared in fixing the gauges. We then do canonical quantization and obtain the critical dimensions for different tensionless strings. We find that among four kinds of tensionless superstrings, the $\mathcal{N}=2$ homogeneous and inhomogeneous doublet tensionless superstrings have the same critical dimension as the usual superstrings. Taking the BMS $bc$ and $βγ$ ghosts as new types of BMS free field theories, we find that their enhanced underlying symmetries are generated by BMS-Kac-Moody algebras, with the Kac-Moody subalgebras being built from a three-dimensional non-abelian and non-semi-simple Lie algebra.

hep-th↗
arXiv2023

Resource Constrained Model Compression via Minimax Optimization for Spiking Neural Networks

Brain-inspired Spiking Neural Networks (SNNs) have the characteristics of event-driven and high energy-efficient, which are different from traditional Artificial Neural Networks (ANNs) when deployed on edge devices such as neuromorphic chips. Most previous work focuses on SNNs training strategies to improve model performance and brings larger and deeper network architectures. It is difficult to deploy these complex networks on resource-limited edge devices directly. To meet such demand, people compress SNNs very cautiously to balance the performance and the computation efficiency. Existing compression methods either iteratively pruned SNNs using weights norm magnitude or formulated the problem as a sparse learning optimization. We propose an improved end-to-end Minimax optimization method for this sparse learning problem to better balance the model performance and the computation efficiency. We also demonstrate that jointly applying compression and finetuning on SNNs is better than sequentially, especially for extreme compression ratios. The compressed SNN models achieved state-of-the-art (SOTA) performance on various benchmark datasets and architectures. Our code is available at https://github.com/chenjallen/Resource-Constrained-Compression-on-SNN.

cs.CV↗
arXiv2023

One-stage Low-resolution Text Recognition with High-resolution Knowledge Transfer

Recognizing characters from low-resolution (LR) text images poses a significant challenge due to the information deficiency as well as the noise and blur in low-quality images. Current solutions for low-resolution text recognition (LTR) typically rely on a two-stage pipeline that involves super-resolution as the first stage followed by the second-stage recognition. Although this pipeline is straightforward and intuitive, it has to use an additional super-resolution network, which causes inefficiencies during training and testing. Moreover, the recognition accuracy of the second stage heavily depends on the reconstruction quality of the first stage, causing ineffectiveness. In this work, we attempt to address these challenges from a novel perspective: adapting the recognizer to low-resolution inputs by transferring the knowledge from the high-resolution. Guided by this idea, we propose an efficient and effective knowledge distillation framework to achieve multi-level knowledge transfer. Specifically, the visual focus loss is proposed to extract the character position knowledge with resolution gap reduction and character region focus, the semantic contrastive loss is employed to exploit the contextual semantic knowledge with contrastive learning, and the soft logits loss facilitates both local word-level and global sequence-level learning from the soft teacher label. Extensive experiments show that the proposed one-stage pipeline significantly outperforms super-resolution based two-stage frameworks in terms of effectiveness and efficiency, accompanied by favorable robustness. Code is available at https://github.com/csguoh/KD-LTR.

cs.CV↗
arXiv2023

An Adaptive Model Ensemble Adversarial Attack for Boosting Adversarial Transferability

While the transferability property of adversarial examples allows the adversary to perform black-box attacks (i.e., the attacker has no knowledge about the target model), the transfer-based adversarial attacks have gained great attention. Previous works mostly study gradient variation or image transformations to amplify the distortion on critical parts of inputs. These methods can work on transferring across models with limited differences, i.e., from CNNs to CNNs, but always fail in transferring across models with wide differences, such as from CNNs to ViTs. Alternatively, model ensemble adversarial attacks are proposed to fuse outputs from surrogate models with diverse architectures to get an ensemble loss, making the generated adversarial example more likely to transfer to other models as it can fool multiple models concurrently. However, existing ensemble attacks simply fuse the outputs of the surrogate models evenly, thus are not efficacious to capture and amplify the intrinsic transfer information of adversarial examples. In this paper, we propose an adaptive ensemble attack, dubbed AdaEA, to adaptively control the fusion of the outputs from each model, via monitoring the discrepancy ratio of their contributions towards the adversarial objective. Furthermore, an extra disparity-reduced filter is introduced to further synchronize the update direction. As a result, we achieve considerable improvement over the existing ensemble attacks on various datasets, and the proposed AdaEA can also boost existing transfer-based attacks, which further demonstrates its efficacy and versatility.

cs.CV↗
arXiv2023

Next-Generation Comprehensive Data-Driven Models of Solar Eruptive Events

Solar flares and coronal mass ejections are interrelated phenomena that together are known as solar eruptive events. These are the main drivers of space weather and understanding their origins is a primary goal of Heliophysics. In this white paper, we advocate for the allocation of sufficient resources to bring together experts in observations and modeling to construct and test next generation data-driven models of solar eruptive events. We identify the key components necessary for constructing comprehensive end-to-end models including global scale 3D MHD resolving magnetic field evolution and reconnection, small scale simulations of particle acceleration in reconnection exhausts, kinetic scale transport of flare-accelerated particles into the lower solar atmosphere, and the radiative and hydrodynamics responses of the solar atmosphere to flare heating. Using this modeling framework, long-standing questions regarding how solar eruptive events release energy, accelerate particles, and heat plasma can be explored. To address open questions in solar flare physics, we recommend that NASA and NSF provide sufficient research and analysis funds to bring together a large body of researchers and numerical tools to tackle the end-to-end modeling framework that we outline. Current dedicated theory and modeling funding programs are relatively small scale and infrequent; funding agencies must recognize that modern space physics demands the use of both observations and modeling to make rapid progress.

astro-ph.IM↗
arXiv2023

Instance-aware Dynamic Prompt Tuning for Pre-trained Point Cloud Models

Pre-trained point cloud models have found extensive applications in 3D understanding tasks like object classification and part segmentation. However, the prevailing strategy of full fine-tuning in downstream tasks leads to large per-task storage overhead for model parameters, which limits the efficiency when applying large-scale pre-trained models. Inspired by the recent success of visual prompt tuning (VPT), this paper attempts to explore prompt tuning on pre-trained point cloud models, to pursue an elegant balance between performance and parameter efficiency. We find while instance-agnostic static prompting, e.g. VPT, shows some efficacy in downstream transfer, it is vulnerable to the distribution diversity caused by various types of noises in real-world point cloud data. To conquer this limitation, we propose a novel Instance-aware Dynamic Prompt Tuning (IDPT) strategy for pre-trained point cloud models. The essence of IDPT is to develop a dynamic prompt generation module to perceive semantic prior features of each point cloud instance and generate adaptive prompt tokens to enhance the model's robustness. Notably, extensive experiments demonstrate that IDPT outperforms full fine-tuning in most tasks with a mere 7% of the trainable parameters, providing a promising solution to parameter-efficient learning for pre-trained point cloud models. Code is available at \url{https://github.com/zyh16143998882/ICCV23-IDPT}.

cs.CV↗
arXiv2023

The Solar Origin of an In Situ Type III Radio Burst Event

Solar type III radio bursts are generated by beams of energetic electrons that travel along open magnetic field lines through the corona and into interplanetary space. However, understanding the source of these electrons and how they escape into interplanetary space remains an outstanding topic. Here we report multi-instrument, multi-perspective observations of an interplanetary type III radio burst event shortly after the second perihelion of the Parker Solar Probe (PSP). This event was associated with a solar jet that produced an impulsive microwave burst event recorded by the Expanded Owens Valley Solar Array (EOVSA). The type III burst event also coincided with the detection of enhanced in situ energetic electrons recorded by both PSP at 0.37 AU and WIND at 1 AU, which were located very closely on the Parker spiral longitudinally. The close timing association and magnetic connectivity suggest that the in situ energetic electrons originated from the jet's magnetic reconnection region. Intriguingly, microwave imaging spectroscopy results suggest that the escaping energetic electrons were injected into a large opening angle of about 90 degrees, which is at least nine times broader than the apparent width of the jet spire. Our findings provide an interpretation for the previously reported, longitudinally broad spatial distribution of flare locations associated with prompt energetic electron events and have important implications for understanding the origin and distribution of energetic electrons in the interplanetary space.

astro-ph.SR↗
arXiv2023

Deep Physics-Guided Unrolling Generalization for Compressed Sensing

By absorbing the merits of both the model- and data-driven methods, deep physics-engaged learning scheme achieves high-accuracy and interpretable image reconstruction. It has attracted growing attention and become the mainstream for inverse imaging tasks. Focusing on the image compressed sensing (CS) problem, we find the intrinsic defect of this emerging paradigm, widely implemented by deep algorithm-unrolled networks, in which more plain iterations involving real physics will bring enormous computation cost and long inference time, hindering their practical application. A novel deep $\textbf{P}$hysics-guided un$\textbf{R}$olled recovery $\textbf{L}$earning ($\textbf{PRL}$) framework is proposed by generalizing the traditional iterative recovery model from image domain (ID) to the high-dimensional feature domain (FD). A compact multiscale unrolling architecture is then developed to enhance the network capacity and keep real-time inference speeds. Taking two different perspectives of optimization and range-nullspace decomposition, instead of building an algorithm-specific unrolled network, we provide two implementations: $\textbf{PRL-PGD}$ and $\textbf{PRL-RND}$. Experiments exhibit the significant performance and efficiency leading of PRL networks over other state-of-the-art methods with a large potential for further improvement and real application to other inverse imaging problems or optimization models.

cs.CV↗
arXiv2023

Mega-cities dominate China's urban greening

Trees play a crucial role in urban environments, offering various ecosystem services that contribute to public health and human well-being. China has initiated a range of urban greening policies over the past decades, however, monitoring their impact on urban tree dynamics at a national scale has proven challenging. In this study, we deployed nano-satellites to quantify urban tree coverage in all major Chinese cities larger than 50 km2 in 2010 and 2019. Our findings indicate that approximately 6000 km2 (11%) of urban areas were covered by trees in 2019, and 76% of these cities experienced an increase in tree cover compared to 2010. Notably, the increase in tree cover in mega-cities such as Beijing, and Shanghai was approximately twice as large as in most other cities (7.69% vs 3.94%). The study employs a data-driven approach towards assessing urban tree cover changes in relation to greening policies, showing clear signs of tree cover increases but also suggesting an uneven implementation primarily benefiting a few mega-cities.

cs.CV↗