Search

As large language models (LLMs) are widely applied across various fields, model compression has become increasingly crucial for reducing costs and improving inference efficiency. Post-training pruning is a promising method that does not require resource-intensive iterative training and only needs a small amount of calibration data to assess the importance of parameters. Previous research has primarily focused on designing advanced pruning methods, while different calibration data's impact on pruning performance still lacks systematical exploration. We fill this blank and surprisingly observe that the effects of calibration data even value more than designing advanced pruning strategies, especially for high sparsity. Our preliminary exploration also discloses that using calibration data similar to the training data can yield better performance. As pre-training data is usually inaccessible for advanced LLMs, we further provide a self-generating calibration data synthesis strategy to construct feasible calibration data. We conduct experiments on the recent strong open-source LLMs (e.g., DCLM, and LLaMA-3), and the results show that the proposed method outperforms commonly used calibration data and can effectively enhance strong pruning methods (e.g., Wanda, OWL)., Comment: under review

Visual brain decoding aims to decode visual information from human brain activities. Despite the great progress, one critical limitation of current brain decoding research lies in the lack of generalization capability to unseen subjects. Prior works typically focus on decoding brain activity of individuals based on the observation that different subjects exhibit different brain activities, while it remains unclear whether brain decoding can be generalized to unseen subjects. This study aims to answer this question. We first consolidate an image-fMRI dataset consisting of stimulus-image and fMRI-response pairs, involving 177 subjects in the movie-viewing task of the Human Connectome Project (HCP). This dataset allows us to investigate the brain decoding performance with the increase of participants. We then present a learning paradigm that applies uniform processing across all subjects, instead of employing different network heads or tokenizers for individuals as in previous methods, which can accommodate a large number of subjects to explore the generalization capability across different subjects. A series of experiments are conducted and we have the following findings. First, the network exhibits clear generalization capabilities with the increase of training subjects. Second, the generalization capability is common to popular network architectures (MLP, CNN and Transformer). Third, the generalization performance is affected by the similarity between subjects. Our findings reveal the inherent similarities in brain activities across individuals. With the emerging of larger and more comprehensive datasets, it is possible to train a brain decoding foundation model in the future. Codes and models can be found at https://github.com/Xiangtaokong/TGBD.

Metric-based few-shot fine-grained classification has shown promise due to its simplicity and efficiency. However, existing methods often overlook task-level special cases and struggle with accurate category description and irrelevant sample information. To tackle these, we propose TAFD-Net: a task adaptive feature distribution network. It features a task-adaptive component for embedding to capture task-level nuances, an asymmetric metric for calculating feature distribution similarities between query samples and support categories, and a contrastive measure strategy to boost performance. Extensive experiments have been conducted on three datasets and the experimental results show that our proposed algorithm outperforms recent incremental learning algorithms., Comment: 10 pages, 2 figures, conference

The first detailed photometric and spectroscopic analysis of the G-type eclipsing binary KM UMa is presented, which indicates that the system is a short-period detached eclipsing binary. The radial velocity curves were calculated using the cross-correlation function method based on Large Sky Area Multi-Object Fiber Spectroscopic Telescope, Sloan Digital Sky Survey, and our observations, which determined the mass ratio as $q=0.45\ (\pm0.04)$. Based on the light curves from the Transiting Exoplanet Survey Satellite, other survey data, and our multiband observations, the positive and negative O'Connell effects have been detected evolving gradually and alternately over the last 20 yr, which can be explained by the presence of spots on the primary component. A superflare event was detected in the SuperWASP data on 2007 February 28, further indicating that KM UMa is a very active system. We calculated its energy to be $5\times10^{34}$ erg by assuming it occurred on the primary star. Utilizing hundreds of medium-resolution spectra and one low-resolution spectrum, the equivalent width variations of the $H_{\alpha}$ line were calculated, indicating the presence of a 5.21 ($\pm0.67$) yr magnetic activity cycle. The orbital period variations were analyzed using the O-C method, detecting a long-term decrease superimposed with a periodic variation. The amplitude of the cyclic variation is $0.01124\ (\pm0.00004)$ day, with a period of $33.66\ (\pm 0.0012)$ yr, which exceeds the 5.21 yr activity cycle, suggesting that this is more likely attributable to the light travel time effect of a third body. Simultaneously, a visual companion has been detected based on the Gaia astrometric data, indicating that KM UMa is actually in a 2+1+1 hierarchical quadruple system.

This paper reviews published research in the field of computer-aided colorization technology. We argue that the colorization task originates from computer graphics, prospers by introducing computer vision, and tends to the fusion of vision and graphics, so we put forward our taxonomy and organize the whole paper chronologically. We extend the existing reconstruction-based colorization evaluation techniques, considering that aesthetic assessment of colored images should be introduced to ensure that colorization satisfies human visual-related requirements and emotions more closely. We perform the colorization aesthetic assessment on seven representative unconditional colorization models and discuss the difference between our assessment and the existing reconstruction-based metrics. Finally, this paper identifies unresolved issues and proposes fruitful areas for future research and development. Access to the project associated with this survey can be obtained at https://github.com/DanielCho-HK/Colorization.

Two-dimensional (2D) multiferroic heterostructures present a promising platform for advanced spin devices by leveraging the coexisting ferromagnetic (FM) and ferroelectric (FE) orders. Through first-principles calculations and micromagnetic simulations, we reveal non-volatile control of metallicity and topological spin textures in the Cr2Ge2Te6/Hf2Ge2Te6(CGT/HGT) heterostructure. Notably, manipulating ferroelectric polarization in HGT significantly modulates the magnetic anisotropy energy (MAE) and Dzyaloshinskii-Moriya interaction (DMI) of CGT/HGT, reversing the easy magnetization axis from in-plane to out-of-plane. By analyzing the atomic-resolved SOC energy (\Delta Esoc), it is found that the cause of the change comes from the Fert-Levy mechanism. Additionally, this polarization control enables the creation and annihilation of bimerons and skyrmions, with interlayer sliding further altering magnetic ordering. Our findings offer valuable insights into magnetoelectric coupling and spin texture manipulation in 2D magnets, highlighting their potential for next-generation spintronic and memory devices., Comment: 9 pages,5 figures

We present experiments to reproduce the characteristics of core-collapse supernovae with different stellar masses and initial explosion energies in the laboratory. In the experiments, shocks are driven in 1.2 atm and 1.9 atm xenon gas by laser with energy from 1600J to 2800J on the SGIII prototype laser facility. The average shock velocities and shocked densities are obtained from experiments. Experimental results reveal that higher laser energy and lower Xe gas density led to higher shock velocity, and lower Xe gas initial density has a higher compression. Modeling of the experiments using the 2D radiation hydrodynamic codes Icefire shows excellent agreement with the experimental results and gives the temperature. These results will contribute to time-domain astrophysical systems, such as gravitational supernovae, where a strong radiative shock propagates outward from the center of the star after the core collapses., Comment: 7 pages, 2 figures, 1 supplement (8 pages, 3 figures, 2 tables), accepted for publication in Science Bulletin

GRB 230812B, detected by the Gamma-Ray Integrated Detectors (GRID) constellation mission, is an exceptionally bright gamma-ray burst (GRB) with a duration of only 3 seconds. Sitting near the traditional boundary ($\sim$ 2 s) between long and short GRBs, GRB 230812B is notably associated with a supernova (SN), indicating a massive star progenitor. This makes it a rare example of a short-duration GRB resulting from stellar collapse. Our analysis, using a time-evolving synchrotron model, suggests that the burst has an emission radius of approximately $10^{14.5}$~cm. We propose that the short duration of GRB 230812B is due to the combined effects of the central engine's activity time and the time required for the jet to break through the stellar envelope. Our findings provide another case that challenges the conventional view that short-duration GRBs originate exclusively from compact object mergers, demonstrating that a broader range of durations exists for GRBs arising from the collapse of massive stars., Comment: 10 pages, 3 tables, 11 figures

A graph $G$ is conflict-free $k$-edge-colorable if there exists an assignment of $k$ colors to $E(G)$ such that for every edge $e\in E(G)$, there is a color that is assigned to exactly one edge among the closed neighborhood of $e$. The smallest $k$ such that $G$ is conflict-free $k$-edge-colorable is called the conflict-free chromatic index of $G$, denoted $\chi'_{CF}(G)$. D\c{e}bski and Przyby\a{l}o showed that $2\le\chi'_{CF}(T)\le 3$ for every tree $T$ of size at least two. In this paper, we present an algorithm to determine the conflict-free chromatic index of a tree without 2-degree vertices, in time $O(|V(T)|)$. This partially answer a question raised by Kamyczura, Meszka and Przyby\a{l}o.

The paper investigates an inverse problem of recovering the internal source from external temperature measurements in photo-thermal effect. The photo-thermal effect actually involves two physical processes: electromagnetic scattering and heat transfer, described by a nonlinear coupled system of Maxwell's equation and the heat transfer equation. The nonlinear coupling term in the system is represented by the square of the modulus of the electromagnetic (missing the phase information of the electromagnetic field), and the absence of this phase information poses a significant challenge to the reconstruction of the internal source. In addition, the interaction and mutual influence of multiple physical fields, including electric field, magnetic field and temperature field, add to the complexity involved in the inversion of the internal source. Based on the potential theory and asymptotic analysis, we prove that the internal source can be uniquely determined up to sign by the external temperature field. This provides a solid theoretical basis for designing the internal source inversion algorithm and further exploring the theoretical aspects of photo-thermal effect.

The Special Affine Fourier Transformation(SAFT), which generalizes several well-known unitary transformations, has been demonstrated as a valuable tool in signal processing and optics. In this paper, we explore the multivariate dynamical sampling problem in shift-invariant spaces associated with the multi-dimensional SAFT. Specifically, we derive a sufficient and necessary condition under which a function in a shift-invariant space can be stably recovered from its dynamical sampling measurements associated with the multi-dimensional SAFT . We also present a straightforward example to elucidate our main result., Comment: 22 pages, 11 figures

We delve into the rotating black hole solutions in de Rham-Gabadadze-Tolley (dRGT) massive gravity. Employing an analytical approach, we solve the field equations for scenarios both devoid of matter and in the presence of an electromagnetic field. Consequently, we obtain black hole solutions that extend beyond the Kerr-Newman family. These solutions are characterized not only by mass $M$, electric charge $Q$, and angular momentum $a$, but also by the graviton mass term, which introduces the cosmological constant $\Lambda$ and the St\"uckelberg charge $S$ into the black hole parameters. Instead of the St\"uckelberg field $\phi^a$, we utilize the matrix $\gamma^2$ to seek solutions. To derive the St\"uckelberg field $\phi^a$ from a given matrix $\gamma^2$ for axisymmetric metrics, one must relax the constraint on the reference metric $f_{ab} = \eta_{ab}$. These solutions potentially serve as candidates for astrophysical black holes.

In order to promote the development of the next generation of nano-spintronic devices, it is of great significance to tune the freedom of valley in two-dimensional (2D) materials. Here, we propose a mechanism for manipulating the valley and nonlinear Hall effect by the 2D ferroelectric substrate. The monolayer Mn2P2S3Se3 is a robust antiferromagnetic valley polarized semiconductor. Importantly, the valley polarized metal-semiconductor phase transition of Mn2P2S3Se3 can be effectively tuned by switching the ferroelectric polarization of Sc2CO2. We reveal the microscopic mechanism of phase transition, which origins from the charge transfer and band alignment. Additionally, we find that transformed polarization direction of Sc2CO2 flexibly manipulate the Berry curvature dipole. Based on this discovery, we present the detection valley polarized metal-semiconductor transition by the nonlinear Hall effect devices. These findings not only offer a scheme to tune the valley degree of freedom, but also provide promising platform to design the nonlinear Hall effect devices., Comment: 8 pages, 6 figures

We employ combinatorial techniques to present an explicit formula for the coefficients in front of Chern classes involving in the Hattori-Stong integrability conditions. We also give an evenness condition for the signature of stably almost-complex manifolds in terms of Chern numbers. As an application, it can be showed that the signature of a $2n$-dimensional stably almost-complex manifold whose possibly nonzero Chern numbers being $c_n$ and $c_ic_{n-i}$ is even, which particularly rules out the existence of such structure on rational projective planes. Some other related results and remarks are also discussed in this article., Comment: 15 pages

Relation classification (RC) plays a pivotal role in both natural language understanding and knowledge graph completion. It is generally formulated as a task to recognize the relationship between two entities of interest appearing in a free-text sentence. Conventional approaches on RC, regardless of feature engineering or deep learning based, can obtain promising performance on categorizing common types of relation leaving a large proportion of unrecognizable long-tail relations due to insufficient labeled instances for training. In this paper, we consider few-shot learning is of great practical significance to RC and thus improve a modern framework of metric learning for few-shot RC. Specifically, we adopt the large-margin ProtoNet with fine-grained features, expecting they can generalize well on long-tail relations. Extensive experiments were conducted by FewRel, a large-scale supervised few-shot RC dataset, to evaluate our framework: LM-ProtoNet (FGF). The results demonstrate that it can achieve substantial improvements over many baseline approaches., Comment: Accepted by CIKM'19

Baidu runs the largest commercial web search engine in China, serving hundreds of millions of online users every day in response to a great variety of queries. In order to build a high-efficiency sponsored search engine, we used to adopt a three-layer funnel-shaped structure to screen and sort hundreds of ads from billions of ad candidates subject to the requirement of low response latency and the restraints of computing resources. Given a user query, the top matching layer is responsible for providing semantically relevant ad candidates to the next layer, while the ranking layer at the bottom concerns more about business indicators (e.g., CPM, ROI, etc.) of those ads. The clear separation between the matching and ranking objectives results in a lower commercial return. The Mobius project has been established to address this serious issue. It is our first attempt to train the matching layer to consider CPM as an additional optimization objective besides the query-ad relevance, via directly predicting CTR (click-through rate) from billions of query-ad pairs. Specifically, this paper will elaborate on how we adopt active learning to overcome the insufficiency of click history at the matching layer when training our neural click networks offline, and how we use the SOTA ANN search technique for retrieving ads more efficiently (Here ``ANN'' stands for approximate nearest neighbor search). We contribute the solutions to Mobius-V1 as the first version of our next generation query-ad matching system., Comment: Accepted by KDD'19

The IW And phenomenon in cataclysmic variables presents a significant challenge to the accretion disk instability model. Using photometric data from the All-Sky Automated Survey for Supernovae, the Zwicky Transient Facility, and the Transiting Exoplanet Survey Satellite, we identify Karachurin 12 as a new IW And-type object with a cycle period of 35.69(3) days. We also report for the first time that Karachurin 12 is a negative superhump (NSH) system featuring a precessing tilted disk, with precession, orbital, and NSH periods of 4.9588(2) days, 0.3168895(13) days, and 0.2979861(8) days, respectively. Our analysis, using dips as indexes and NSHs as probes, reveals diverse cycle patterns in Karachurin 12, with NSH amplitude varying throughout the cycle. Analyses of NSH behaviour using multiple methods shows that the amplitude decreases with increasing outbursts and increases with weakening outbursts, which we suggest is related to changes in the radius of the accretion disk. The mass transfer outburst model fails to explain the observed NSH amplitude variations, whereas the tilted thermally unstable disk model accounts for most of the observed cycles and dip patterns, particularly the disappearance of the IW And phenomenon when NSHs become undetectable. However, no significant changes in the NSH period were observed. These findings suggest a potential link between the IW And phenomenon and a tilted disk. Further improvements in simulations are needed to better understand the detailed dynamics of these cycles., Comment: 22 pages, 9 figures and 3 tables

The buried interface between the electron transport layer and the perovskite layer suffers from severe interface defects and imperfect energy level alignment. To address this issue, this study employs a multifunctional organic molecule, curcumin, to modify the interface between SnO2 and the perovskite layer. The functional groups on curcumin effectively passivate the defects on both sides of the interface, reducing -OH and oxygen vacancy defects on the SnO2 surface and passivating uncoordinated Pb2+ in the perovskite layer. This results in a more compatible energy level alignment and lower defect density at the interface, enhancing carrier transport across it. Consequently, the devices based on curcumin achieve an impressive champion power conversion efficiency (PCE) of 24.46%, compared to 22.03% for control devices. This work demonstrates a simple, green, hydrophobic, and efficient molecular modification method for the buried interface, laying the foundation for the development of high-performance and stable perovskite solar cells.

Recently, large language and vision models have shown strong performance, but due to high pre-training and fine-tuning costs, research has shifted towards faster training via dataset pruning. Previous methods used sample loss as an evaluation criterion, aiming to select the most "difficult" samples for training. However, when the pruning rate increases, the number of times each sample is trained becomes more evenly distributed, which causes many critical or general samples to not be effectively fitted. We refer to this as Low-Frequency Learning (LFL). In other words, LFL prevents the model from remembering most samples. In our work, we decompose the scoring function of LFL, provide a theoretical explanation for the inefficiency of LFL, and propose adding a memory term to the scoring function to enhance the model's memory capability, along with an approximation of this memory term. Similarly, we explore memory in Self-Supervised Learning (SSL), marking the first discussion on SSL memory. Using contrastive learning, we derive the memory term both theoretically and experimentally. Finally, we propose Enhance Memory Pruning (EMP), which addresses the issue of insufficient memory under high pruning rates by enhancing the model's memory of data, thereby improving its performance. We evaluated the performance of EMP in tasks such as image classification, natural language understanding, and model pre-training. The results show that EMP can improve model performance under extreme pruning rates. For example, in the CIFAR100-ResNet50 pre-training task, with 70\% pruning, EMP outperforms current methods by 2.2\%.

Video Shadow Detection (VSD) aims to detect the shadow masks with frame sequence. Existing works suffer from inefficient temporal learning. Moreover, few works address the VSD problem by considering the characteristic (i.e., boundary) of shadow. Motivated by this, we propose a Timeline and Boundary Guided Diffusion (TBGDiff) network for VSD where we take account of the past-future temporal guidance and boundary information jointly. In detail, we design a Dual Scale Aggregation (DSA) module for better temporal understanding by rethinking the affinity of the long-term and short-term frames for the clipped video. Next, we introduce Shadow Boundary Aware Attention (SBAA) to utilize the edge contexts for capturing the characteristics of shadows. Moreover, we are the first to introduce the Diffusion model for VSD in which we explore a Space-Time Encoded Embedding (STEE) to inject the temporal guidance for Diffusion to conduct shadow detection. Benefiting from these designs, our model can not only capture the temporal information but also the shadow property. Extensive experiments show that the performance of our approach overtakes the state-of-the-art methods, verifying the effectiveness of our components. We release the codes, weights, and results at \url{https://github.com/haipengzhou856/TBGDiff}., Comment: ACM MM2024

The standard external shock model in the thin-shell scenario predicts an onset bump in the early optical afterglow light curves of gamma-ray bursts (GRBs). We collect such a textbook-version light curve sample of $30$ GRBs, and derive the jet properties from our joint fit to their X-ray and optical afterglow light curves. It is found that the distributions of the isotropic initial Lorentz factors ($\Gamma_0$), the deceleration radii ($R_{\rm dec}$), and the magnetic field strength ($B_0$) are log-normal, but the distributions of the isotropic kinetic energy ($E_{\rm k, iso}$), medium density ($n_{0}$), and the magnetization parameter ($\sigma_{B}\equiv\epsilon_B/\epsilon_e$) are tentatively bimodal. A tight $R_{\rm dec}\mbox{-}B_{0}\mbox{-}\sigma_{B}$ relation is found. It infers a universal $\epsilon_e E_{\rm k,iso}$ among bursts, plausibly supporting the previous argument of a universal GRB radiation energy among GRBs. A jet break is required for modeling the light curves of $26$ GRBs. The distributions of the jet opening angles and the jet-corrected kinetic energies log-normally center at $\log \theta_{\rm j,c}/{\rm rad}=-1.51$ (standard deviation $\sigma=0.27$) and $\log (E_{\rm k, j,c}/{\rm erg})=51.78$ ($\sigma=0.54$), respectively. Those GRBs ($19$ GRBs), whose prompt gamma-ray emission is well estimated with broad energy-band observations, satisfy the previously discovered $L_{\rm \gamma, p, iso}-E_{\rm p,z}-\Gamma_{0}$ relation, and their gamma-ray radiation efficiencies log-normally distribute in the range from $0.04\%$ to $10\%$ with a central value of $0.42\%$. Such a low efficiency favors the baryonic fireball model, and the distribution of their baryon mass loading in the GRB ejecta log-normally centers at $\log (M_{\rm fb,c}/M_{\rm sun})=-5$ ($\sigma=0.75$)., Comment: 12 pages, 4 tables, 9 figures. Publication in the Astrophysics Journal (in press)

Pruning is a critical strategy for compressing trained large language models (LLMs), aiming at substantial memory conservation and computational acceleration without compromising performance. However, existing pruning methods often necessitate inefficient retraining for billion-scale LLMs or rely on heuristic methods such as the optimal brain surgeon framework, which degrade performance. In this paper, we introduce FISTAPruner, the first post-training pruner based on convex optimization models and algorithms. Specifically, we propose a convex optimization model incorporating $\ell_1$ norm to induce sparsity and utilize the FISTA solver for optimization. FISTAPruner incorporates an intra-layer cumulative error correction mechanism and supports parallel pruning. We comprehensively evaluate FISTAPruner on models such as OPT, LLaMA, LLaMA-2, and LLaMA-3 with 125M to 70B parameters under unstructured and 2:4 semi-structured sparsity, demonstrating superior performance over existing state-of-the-art methods across various language benchmarks.

We present our optical observations and multi-wavelength analysis of the GRB\,200613A detected by \texttt{Fermi} satellite. Time-resolved spectral analysis of the prompt $\gamma$-ray emission was conducted utilizing the Bayesian block method to determine statistically optimal time bins. Based on the Bayesian Information Criterion (BIC), the data generally favor the Band+Blackbody (short as BB) model. We speculate that the main Band component comes from the Blandford-Znajek mechanism, while the additional BB component comes from the neutrino annihilation process. The BB component becomes significant for a low-spin, high-accretion rate black hole central engine, as evidenced by our model comparison with the data. The afterglow light curve exhibits typical power-law decay, and its behavior can be explained by the collision between the ejecta and constant interstellar medium (ISM). Model fitting yields the following parameters: $E_{K,iso} = (2.04^{+11.8}_{-1.50})\times 10^{53}$ erg, $\Gamma_0=354^{+578}_{-217}$, $p=2.09^{+0.02}_{-0.03}$, $n_{18}=(2.04^{+9.71}_{-1.87})\times 10^{2}$ cm$^{-3}$, $\theta_j=24.0^{+6.50}_{-5.54}$ degree, $\epsilon_e=1.66^{+4.09}_{-1.39})\times 10^{-1}$ and $\epsilon_B=(7.76^{+48.5}_{-5.9})\times 10^{-6}$. In addition, we employed the public Python package \texttt{Prospector} perform a spectral energy distribution (SED) modeling of the host galaxy. The results suggest that the host galaxy is a massive galaxy ($\log(M_\ast / M_\odot)=11.75^{+0.10}_{-0.09}$) with moderate star formation rate ($\mbox{SFR}=22.58^{+13.63}_{-7.22} M_{\odot}$/yr). This SFR is consistent with the SFR of $\sim 34.2 M_{\odot}$ yr$^{-1}$ derived from the [OII] emission line in the observed spectrum., Comment: 30 pages, 16 figures, accepted by ApJ

With the widespread adoption of Large Language Models (LLMs), concerns about potential misuse have emerged. To this end, watermarking has been adapted to LLM, enabling a simple and effective way to detect and monitor generated text. However, while the existing methods can differentiate between watermarked and unwatermarked text with high accuracy, they often face a trade-off between the quality of the generated text and the effectiveness of the watermarking process. In this work, we present a novel type of LLM watermark, Sparse Watermark, which aims to mitigate this trade-off by applying watermarks to a small subset of generated tokens distributed across the text. The key strategy involves anchoring watermarked tokens to words that have specific Part-of-Speech (POS) tags. Our experimental results demonstrate that the proposed watermarking scheme achieves high detectability while generating text that outperforms previous LLM watermarking methods in quality across various tasks

We propose a quantum beam splitter (QBS) with tunable reflection and transmission coefficients. More importantly, our device based on a Hermitian parity-time ($\mathcal{PT}$) symmetric system enables the generation and manipulation of asymmetric quantum coherence of the output photons. For the interference of two weak coherent-state inputs, our QBS can produce anti-bunched photons from one output port and bunched photons from the other, showcasing high parity asymmetry and strong coherence control capabilities. Beyond the Hong-Ou-Mandel effect, perfect photon blockade with vanishing $g^{(2)}(0)$ is achievable in two-photon interference. These striking effects of the QBS fundamentally arise from the parity-symmetry-breaking interaction and the quantum interference between the photon scattering channels. Our results could inspire novel applications and the development of innovative photonic devices for the manipulation of weak quantum light., Comment: The document consists of 6 figures and spans 22 pages, including detailed appendices

M-type stars are the ones that flare most frequently, but how big their maximum flare energy can reach is still unknown. We present 163 flares from 162 individual M2 through L1-type stars that triggered the GWAC, with flare energies ranging from $10^{32.2}$ to $10^{36.4}$ erg . The flare amplitudes range from $\triangle G = 0.84$ to $\sim 10$ mag. Flare energy increases with stellar surface temperature ($T_{\rm eff}$) but both $\triangle G$ and equivalent duration $\log_{10}(ED)$ seem to be independent of $T_{\rm eff}$. Combining periods detected from light curves of TESS and K2, spectra from LAMOST, SDSS and the 2.16 m Telescope, and the Gaia DR3 data, we found that these GWAC flare stars are young. For the stars that have spectra, we found that these stars are in or very near to the saturation region, and $\log_{10}(L_{\rm H\alpha}/L_{\rm bol})$ is lower for M7-L1 stars than for M2-M6 stars. We also studied the relation between GWAC flare bolometric energy $E_{\rm bol}$ and stellar hemispherical area $S$, and found that $\log_{10}E_{\rm bol}$ (in erg) increases with increasing $S$ (in cm$^2$), and the maximum flare energy $\log_{10}E_{\rm bol, max} \geqslant \log_{10}S + 14.25$. For M7-L1 stars, there seem to be other factors limiting their maximum flare energies in addition to stellar hemispherical area., Comment: 18 pages, 11 figures, 4 tables

Tilted disk precession exists in different objects. Negative superhumps (NSHs) in cataclysmic variable stars (CVs) are believed to arise from the interaction between the reverse precession of a tilted disk and the streams from the secondary star.Utilizing TESS photometry, we present a comprehensive investigation into the tilted disk precession and NSHs in the dwarf nova (DN) HS 2325+8205, employing eclipse minima, eclipse depths, NSH frequencies, and NSH amplitudes and the correlation between them as the windows. We identified NSHs with a period of 0.185671(17) days in HS 2325+8205. The NSH frequency exhibits variability with a period of 3.943(9) days, akin to the tilted disk precession period validated in novae-like stars (NLs, SDSS J0812) and intermediate polars (IPs, TV Col).The O-C of eclipse minima were similarly found to vary cyclically in period 4.135(5) days, characterized by a faster rise than fall. Furthermore, the NSH amplitude exhibits complex and diverse variations, which may be linked to changes in the disk radius, mass transfer rate, and the apparent area of the hot spot. For the first time in DNe, we observe bi-periodic variations in eclipse depth (P1= 4.131(4) d and P2= 2.065(2) d ~ Pprec/2), resembling those seen in IPs, suggesting that variations with P2 are not attributable to an accretion curtain, as previously suspected. Moreover, NSH amplitude and eclipse depth decrease with increasing NSH frequency, while NSH amplitude correlates positively with eclipse depth.These complex variations observed across multiple observational windows provide substantial evidence for understanding of tilted disk precession and NSHs., Comment: 19 pages, 10 figures and 3 tables; accepted for publication in ApJ

SZ Lyn is an astrometric binary consisting of a $\delta$ Scuti star and an unseen component. The two stars around each other in an almost circular orbit (e=0.186) with a period of about 3.32 years. We aim to explore the physical properties of the two components using the radial velocity method and asteroseismology. Our results show that the $\delta$ Scuti variable is a post-main sequence star with a helium core ($M_{\rm He}$= 0.1671$^{+0.0028}_{-0.0045}$ M$_{\odot}$ and $R_{\rm He}$= 0.1414$^{+0.0035}_{-0.0020}$ R$_{\odot}$) surrounded by a burning hydrogen shell; its fundamental parameters is accurately calculated to be $M$=1.82$^{+0.04}_{-0.02}$ M$_{\odot}$, $R$=2.91$^{+0.02}_{-0.02}$ R$_{\odot}$, $L$=17.07$^{+1.28}_{-2.61}$ L$_{\odot}$, and $X_c$=0.078$^{+0.021}_{-0.004}$; the mass of the invisible object is thus estimated to be $M_{\rm co}$=1.68 $^{+0.62}_{-0.60}$ M$_{\odot}$. Combined with the low-resolution LAMOST spectrum, we suggest that the unseen object is an neutron star candidate. Our work demonstrates the methodology of using the radial velocity method and asteroseismology to detect the compact object in such an astrometric binary., Comment: In preparation

We investigate the stationary, spherically symmetric accretion of a degenerate relativistic Fermi gas onto a Reissner-Nordstr\"{o}m black hole. The accretion theory is based on the Boyer-Lindquist coordinates and the Fermi gas follows Fermi-Dirac statistics at infinity. We have derived the expression for the particle current density, the stress energy-momentum tensor, and three accretion rates. As the charged particle falls into the black hole, both the mass and the charge of the black hole increase. Consequently, the mass accretion rate and charge accretion rate are proportional to the particle accretion rate. We have also provided analytical results at infinity and numerical results within a finite range for these quantities. Our results indicate that the accretion rate decreases as the charge of the black hole increases, suggests that the presence of naked singularities is avoided in black hole accretion theory. In this paper, we also discovered that the accretion model of Vlasov gas behaves as an anisotropic fluid containing two perfect-fluid components. One component represents the isotropic fluid of Fermi gas, while the other represents a null fluid. When using the Boyer-Lindquist coordinate system, we observed that the contribution from the null fluid persists even at infinity, which led to the radial pressure always smaller than the tangential pressure. Therefore, it's not appropriate to treat the accretion model as a perfect fluid at infinity.

Last-iterate behaviors of learning algorithms in repeated two-player zero-sum games have been extensively studied due to their wide applications in machine learning and related tasks. Typical algorithms that exhibit the last-iterate convergence property include optimistic and extra-gradient methods. However, most existing results establish these properties under the assumption that the game is time-independent. Recently, (Feng et al, 2023) studied the last-iterate behaviors of optimistic and extra-gradient methods in games with a time-varying payoff matrix, and proved that in an unconstrained periodic game, extra-gradient method converges to the equilibrium while optimistic method diverges. This finding challenges the conventional wisdom that these two methods are expected to behave similarly as they do in time-independent games. However, compared to unconstrained games, games with constrains are more common both in practical and theoretical studies. In this paper, we investigate the last-iterate behaviors of optimistic and extra-gradient methods in the constrained periodic games, demonstrating that similar separation results for last-iterate convergence also hold in this setting., Comment: Accepted for UAI 2024

We reported a cyclic variation of $O-C$ diagram with a semi-amplitude of 0.0033 days and a period of 1.05 years for the pulsating eclipsing binary HZ Dra. The cyclic variation can be explained by the light travel-time effect via the presence of a close-in third body orbiting around HZ Dra in an elliptical orbit with a maximum semi-major axis of 0.92 au. Based on the W-D code, the contribution of the third light to the total system is determined to be 29 $\%$, which is in agreement with the estimated value. Our light curve modelling indicates an evolving hot and cool spot on the surface of the primary and secondary components, respectively. Their positions are roughly symmetrical to the inner Lagrangian point L1, which could be used to explain the variation in the O$^{'}$Connell effect. Our frequency analysis detects 1 radial p-mode, 7 non-radial p-modes and 1 non-radial g-mode. In addition, a total of 6 multiplets are identified, spaced by the orbital frequency, which can be explained as a tidally split mode caused by the equilibrium tides of the close binary system with a circular orbit. These pulsating features suggest that the primary of HZ Dra is a $\delta$ Scuti star, pulsating in both p- and g-mode and influenced by tidal forces., Comment: 15 pages, 11 figures. The first version is Under Review in ApJ

As a defense strategy against adversarial attacks, adversarial detection aims to identify and filter out adversarial data from the data flow based on discrepancies in distribution and noise patterns between natural and adversarial data. Although previous detection methods achieve high performance in detecting gradient-based adversarial attacks, new attacks based on generative models with imbalanced and anisotropic noise patterns evade detection. Even worse, existing techniques either necessitate access to attack data before deploying a defense or incur a significant time cost for inference, rendering them impractical for defending against newly emerging attacks that are unseen by defenders. In this paper, we explore the proximity relationship between adversarial noise distributions and demonstrate the existence of an open covering for them. By learning to distinguish this open covering from the distribution of natural data, we can develop a detector with strong generalization capabilities against all types of adversarial attacks. Based on this insight, we heuristically propose Perturbation Forgery, which includes noise distribution perturbation, sparse mask generation, and pseudo-adversarial data production, to train an adversarial detector capable of detecting unseen gradient-based, generative-model-based, and physical adversarial attacks, while remaining agnostic to any specific models. Comprehensive experiments conducted on multiple general and facial datasets, with a wide spectrum of attacks, validate the strong generalization of our method.

We present photometric and spectroscopic observations and analysis of the type IIb supernova (SN) SN 2019tua, which exhibits multiple bumps in its declining light curves between 40 and 65 days after discovery. SN 2019tua shows a time to peak of about 25 days similar to other type IIb SNe. Our observations indicate a decrease in its brightness of about 1 magnitude in the 60 days after the peak. At about days 50, and 60, its multiband light curves exhibit bumpy behavior. The complex luminosity evolution of SN 2019tua could not be well modeled with a single currently popular energy source model, e.g., radioactive decay of $^{56}$Ni, magnetar, interaction between the ejecta and a circumstellar shell. Even though the magnetar model has a smaller $\chi^2 / \text{dof}$ value, the complex changes in SN 2019tua's brightness suggest that more than one physical process might be involved. We propose a hybrid CSM interaction plus $^{56}$Ni model to explain the bolometric light curve (LC) of SN 2019tua. The fitting results show that the ejecta mass $M_{\rm ej} \approx 2.4~M_\odot$, the total CSM mass $M_{\rm CSM} \approx 1.0~M_\odot$, and the $^{56}$Ni mass $M_{\rm Ni} \approx 0.4~M_\odot$. The total kinetic energy of the ejecta is $E_k\approx 0.5 \times 10^{51}\rm~erg$. Pre-existing multiple shells suggest that the progenitor of SN 2019tua experienced mass ejections within approximately $\sim6 - 44$ years prior to the explosion., Comment: Accepted for publication in ApJ. 24 pages, 17 figures, 6 tables

Large language models have demonstrated strong performance in recent years, but the high cost of training drives the need for efficient methods to compress dataset sizes. We propose TED pruning, a method that addresses the challenge of overfitting under high pruning ratios by quantifying the model's ability to improve performance on pruned data while fitting retained data, known as Internal Generalization (IG). TED uses an optimization objective based on Internal Generalization Distance (IGD), measuring changes in IG before and after pruning to align with true generalization performance and achieve implicit regularization. The IGD optimization objective was verified to allow the model to achieve the smallest upper bound on generalization error. The impact of small mask fluctuations on IG is studied through masks and Taylor approximation, and fast estimation of IGD is enabled. In analyzing continuous training dynamics, the prior effect of IGD is validated, and a progressive pruning strategy is proposed. Experiments on image classification, natural language understanding, and large language model fine-tuning show TED achieves lossless performance with 60-70\% of the data. Upon acceptance, our code will be made publicly available.

Quantum electrodynamics (QED) deals with the relativistic interaction of bosonic gauge fields and fermionic charged particles. In QED, global conservation laws of angular momentum for light-matter interactions are well-known. However, local conservation laws, i.e. the conservation law of angular momentum at every point in space, remain unexplored. Here, we use the QED Lagrangian and Noether's theorem to derive a new local conservation law of angular momentum for Dirac-Maxwell fields in the form of the continuity relation for linear momentum. We separate this local conservation law into four coupled motion equations for spin and orbital angular momentum (OAM) densities. We introduce a helicity current tensor, OAM current tensor, and spin-orbit torque in the motion equations to shed light on on the local dynamics of spin-OAM interaction and angular momentum exchange between Maxwell-Dirac fields. We elucidate how our results translate to classical electrodynamics using the example of plane wave interference as well as a dual-mode optical fiber. Our results shine light on phenomena related to the spin of gauge bosons.

The stability of perovskite solar cells is closely related to the defects in perovskite crystals, and there are a large number of crystal defects in the perovskite thin films prepared by the solution method, which is not conducive to the commercial production of PSCs. In this study, resveratrol(RES), a green natural antioxidant abundant in knotweed and grape leaves, was introduced into perovskite films to passivate the defect. RES achieves defect passivation by interacting with uncoordinated Pb2+ in perovskite films. The results show that the quality of the perovskite film is significantly improved, and the energy level structure of the device is optimized, and the power conversion efficiency of the device is increased from 21.62% to 23.44%. In addition, RES can hinder the degradation of perovskite structures by O2- and CO2- free radicals, and the device retained 88% of its initial PCE after over 1000 hours in pure oxygen environment. The device retains 91% of the initial PCE after more than 1000 hours at 25{\deg}C and 50+5% relative humidity. This work provides a strategy for the use of natural and environmentally friendly additives to improve the efficiency and stability of devices, and provides an idea for the development of efficient, stable and environmentally friendly PSCs.

Recent self-training techniques have shown notable improvements in unsupervised domain adaptation for 3D object detection (3D UDA). These techniques typically select pseudo labels, i.e., 3D boxes, to supervise models for the target domain. However, this selection process inevitably introduces unreliable 3D boxes, in which 3D points cannot be definitively assigned as foreground or background. Previous techniques mitigate this by reweighting these boxes as pseudo labels, but these boxes can still poison the training process. To resolve this problem, in this paper, we propose a novel pseudo label refinery framework. Specifically, in the selection process, to improve the reliability of pseudo boxes, we propose a complementary augmentation strategy. This strategy involves either removing all points within an unreliable box or replacing it with a high-confidence box. Moreover, the point numbers of instances in high-beam datasets are considerably higher than those in low-beam datasets, also degrading the quality of pseudo labels during the training process. We alleviate this issue by generating additional proposals and aligning RoI features across different domains. Experimental results demonstrate that our method effectively enhances the quality of pseudo labels and consistently surpasses the state-of-the-art methods on six autonomous driving benchmarks. Code will be available at https://github.com/Zhanwei-Z/PERE., Comment: Accepted by CVPR2024

We study electromagnetic form factors of protons in proton-nucleus scattering via analysing of experimental cross-sections of accompanying bremsstrahlung photons. A new bremsstrahlung model for proton-nucleus scattering is developed, where a main focus is given on incoherent bremsstrahlung that has not been considered previously. In analysis we choose experimental bremsstrahlung data of $p$ + $^{197}$Au scattering at proton beam energy of 190 MeV obtained by TAPS collaboration. We find the following. (1) Inclusion of incoherent emission to calculations improves agreements with experimental data essentially, contribution of incoherent bremsstrahlung is essentially larger than coherent one. (2) Inclusion of form factors of the scattered proton improves agreement with experimental data in comparison with calculations with coherent and incoherent contributions without form factors. (3) Sensitivity of model in study of form factors of the scattered proton is high. This demonstrates a new opportunity to study internal structure of protons under influence of nuclear forces in nuclear scattering., Comment: 32 pages, 4 captured figures

In this paper, we extend Chandrasekhar's method of calculating rotating black holes into $f(R)$ theory. We consider the Ricci scalar is a constant and derive the Kerr and Kerr-Ads metric by using the analytical mathematical method. Suppose that the spacetime is a 4-dimensional Riemannian manifold with a general stationary axisymmetric metric, we calculate Cartan's equation of structure and derive the Einstein tensor. In order to reduce the solving difficulty, we fix the gauge freedom to transform the metric into a more symmetric form. We solve the field equations in the two cases of the Ricci scalar $R=0$ and $R\neq 0$. In the case of $R=0$, the Ernst's equations are derived. We give the elementary solution of Ernst's equations and show the way to obtain more solutions including Kerr metric. In the case of $R\neq 0$, we reasonably assume that the solution to the equations consists of two parts: the first is Kerr part and the second is introduced by the Ricci scalar. Giving solution to the second part and combining the two parts, we obtain the Kerr-Ads metric. The calculations are carried out in a general $f(R)$ theory, indicating the Kerr and Kerr-Ads black holes exist widely in general $f(R)$ models. Furthermore, the whole solving process can be treated as a standard calculation procedure to obtain rotating black holes, which can be applied to other modified gravities.

The total angular momentum of light has received attention for its application in a variety of phenomena such as optical communication, optical forces and sensing. However, the quantum behavior including the commutation relations have been relatively less explored. Here, we derive the correct commutation relation for the total angular momentum of light using both relativistic and non-relativistic approaches. An important outcome of our work is the proof that the widely-assumed quantum commutation relation for the total observable angular momentum of light is fundamentally incorrect. Our work will motivate experiments and leads to new insight on the quantum behavior of the angular momentum of light.

As deep learning models exponentially increase in size, optimizers such as Adam encounter significant memory consumption challenges due to the storage of first and second moment data. Current memory-efficient methods like Adafactor and CAME often compromise accuracy with their matrix factorization techniques. Addressing this, we introduce Adapprox, a novel approach that employs randomized low-rank matrix approximation for a more effective and accurate approximation of Adam's second moment. Adapprox features an adaptive rank selection mechanism, finely balancing accuracy and memory efficiency, and includes an optional cosine similarity guidance strategy to enhance stability and expedite convergence. In GPT-2 training and downstream tasks, Adapprox surpasses AdamW by achieving 34.5% to 49.9% and 33.8% to 49.9% memory savings for the 117M and 345M models, respectively, with the first moment enabled, and further increases these savings without the first moment. Besides, it enhances convergence speed and improves downstream task performance relative to its counterparts.

This paper addresses the problem of detecting a moving target embedded in Gaussian noise with an unknown covariance matrix for frequency diverse array multiple-input multiple-output (FDA-MIMO) radar. To end it, assume that obtaining a set of training data is available. Moreover, we propose three adaptive detectors in accordance with the one-step generalized likelihood ratio test (GLRT), two-step GLRT, and Rao criteria, namely OGLRT, TGLRT, and Rao. The LH adaptive matched filter (LHAMF) detector is also introduced when decomposing the Rao test. Next, all provided detectors have constant false alarm rate (CFAR) properties against the covariance matrix. Besides, the closed-form expressions for false alarm probability (PFA) and detection probability (PD) are derived. Finally, this paper substantiates the correctness of the aforementioned algorithms through numerical simulations.

Pruning before training enables the deployment of neural networks on smart devices. By retaining weights conducive to generalization, pruned networks can be accommodated on resource-constrained smart devices. It is commonly held that the distance on weight norms between the initialized and the fully-trained networks correlates with generalization performance. However, as we have uncovered, inconsistency between this metric and generalization during training processes, which poses an obstacle to determine the pruned structures on smart devices in advance. In this paper, we introduce the concept of the learning gap, emphasizing its accurate correlation with generalization. Experiments show that the learning gap, in the form of feature maps from the penultimate layer of networks, aligns with variations of generalization performance. We propose a novel learning framework, LNPT, which enables mature networks on the cloud to provide online guidance for network pruning and learning on smart devices with unlabeled data. Our results demonstrate the superiority of this approach over supervised training., Comment: 8 pages,7 figures

Large-scale Transformer models (TM) have demonstrated outstanding performance across various tasks. However, their considerable parameter size restricts their applicability, particularly on mobile devices. Due to the dynamic and intricate nature of gradients on TM compared to Convolutional Neural Networks, commonly used pruning methods tend to retain weights with larger gradient noise. This results in pruned models that are sensitive to sparsity and datasets, exhibiting suboptimal performance. Symbolic Descent (SD) is a general approach for training and fine-tuning TM. In this paper, we attempt to describe the noisy batch gradient sequences on TM through the cumulative process of SD. We utilize this design to dynamically assess the importance scores of weights.SEVEN is introduced by us, which particularly favors weights with consistently high sensitivity, i.e., weights with small gradient noise. These weights are tended to be preserved by SEVEN. Extensive experiments on various TM in natural language, question-answering, and image classification domains are conducted to validate the effectiveness of SEVEN. The results demonstrate significant improvements of SEVEN in multiple pruning scenarios and across different sparsity levels. Additionally, SEVEN exhibits robust performance under various fine-tuning strategies. The code is publicly available at https://github.com/xiaojinying/SEVEN., Comment: 9 pages,6 figures

The two-dimensional (2D) multiferroic materials have widespread of application prospects in facilitating the integration and miniaturization of nanodevices. However, it is rarely coupling between the magnetic, ferroelectric, and ferrovalley in one 2D material. Here, we propose a mechanism for manipulating magnetism, ferroelectric, and valley polarization by interlayer sliding in 2D bilayer material. Monolayer GdI2 exhibits a ferromagnetic semiconductor with the valley polarization up to 155.5 meV. More interestingly, the magnetism and valley polarization of bilayer GdI2 can be strongly coupled by sliding ferroelectricity, appearing these tunable and reversible. In addition, we uncover the microscopic mechanism of magnetic phase transition by spin Hamiltonian and electron hopping between layers. Our findings offer a new direction for investigating 2D multiferroic in the implication for next-generation electronic, valleytronic, and spintronic devices., Comment: 21 pages, 5 figures Accepted Nano Lett.(2024)

JPEG compression can significantly impair the performance of adversarial face examples, which previous adversarial attacks on face recognition (FR) have not adequately addressed. Considering this challenge, we propose a novel adversarial attack on FR that aims to improve the resistance of adversarial examples against JPEG compression. Specifically, during the iterative process of generating adversarial face examples, we interpolate the adversarial face examples into a smaller size. Then we utilize these interpolated adversarial face examples to create the adversarial examples in the next iteration. Subsequently, we restore the adversarial face examples to their original size by interpolating. Throughout the entire process, our proposed method can smooth the adversarial perturbations, effectively mitigating the presence of high-frequency signals in the crafted adversarial face examples that are typically eliminated by JPEG compression. Our experimental results demonstrate the effectiveness of our proposed method in improving the JPEG-resistance of adversarial face examples.