Xiuwen Li, Jiaxue Chai, Huixian Zhu, Pei Wang
As is well known, crystals have discrete space translational symmetry. It was recently noticed that one-dimensional crystals possibly have discrete Poincaré symmetry, which contains discrete Lorentz and discrete time translational symmetry as well. In this paper, we classify the discrete Poincaré groups on two- and three-dimensional Bravais lattices. They are the candidate symmetry groups of two- or three-dimensional crystals, respectively. The group is determined by an integer generator $g$, and it reduces to the space group of crystals at $g=2$.
Jie Peng, Ji Zhu, Anna Bergamaschi, Wonshik Han, Dong-Young Noh, Jonathan R. Pollack, Pei Wang
In this paper, we propose a new method remMap -- REgularized Multivariate regression for identifying MAster Predictors -- for fitting multivariate response regression models under the high-dimension-low-sample-size setting. remMap is motivated by investigating the regulatory relationships among different biological molecules based on multiple types of high dimensional genomic data. Particularly, we are interested in studying the influence of DNA copy number alterations on RNA transcript levels. For this purpose, we model the dependence of the RNA expression levels on DNA copy numbers through multivariate linear regressions and utilize proper regularizations to deal with the high dimensionality as well as to incorporate desired network structures. Criteria for selecting the tuning parameters are also discussed. The performance of the proposed method is illustrated through extensive simulation studies. Finally, remMap is applied to a breast cancer study, in which genome wide RNA transcript levels and DNA copy numbers were measured for 172 tumor samples. We identify a tran-hub region in cytoband 17q12-q21, whose amplification influences the RNA expression levels of more than 30 unlinked genes. These findings may lead to a better understanding of breast cancer pathology.
Pei Wang, Zhaowei Cai, Hao Yang, Ashwin Swaminathan, R. Manmatha, Stefano Soatto
Existing unified image segmentation models either employ a unified architecture across multiple tasks but use separate weights tailored to each dataset, or apply a single set of weights to multiple datasets but are limited to a single task. In this paper, we introduce the Mixed-Query Transformer (MQ-Former), a unified architecture for multi-task and multi-dataset image segmentation using a single set of weights. To enable this, we propose a mixed query strategy, which can effectively and dynamically accommodate different types of objects without heuristic designs. In addition, the unified architecture allows us to use data augmentation with synthetic masks and captions to further improve model generalization. Experiments demonstrate that MQ-Former can not only effectively handle multiple segmentation datasets and tasks compared to specialized state-of-the-art models with competitive performance, but also generalize better to open-set segmentation tasks, evidenced by over 7 points higher performance than the prior art on the open-vocabulary SeginW benchmark.
Pei Wang
Dec 28, 2021·quant-ph·PDF We develop an action formulation of stochastic dynamics in the Hilbert space. By generalizing the Wiener process into 1+3-dimensional spacetime, we define a Lorentz-invariant random field. By coupling the random to quantum fields, we obtain a random-number action which has the statistical spacetime translation and Lorentz symmetries. The canonical quantization of the theory results in a Lorentz-invariant equation of motion for the state vector or density matrix. We derive the path integral formula of $S$-matrix and the diagrammatic rules for both the stochastic free field theory and stochastic $φ^4$-theory. The Lorentz invariance of the random $S$-matrix is strictly proved. We then develop a diagrammatic technique for calculating the density matrix. Without interaction, we obtain the exact $S$-matrix and density matrix. With interaction, we prove a simple relation between the density matrices of stochastic and conventional $φ^4$-theory. Our formalism leads to an ultraviolet divergence which has the similar origin as that in QFT. The divergence is canceled by renormalizing the coupling strength to random field. We prove that the stochastic QFT is renormalizable even in the presence of interaction. In the models with a linear coupling between random and quantum fields, the random field excites particles out of the vacuum, driving the universe towards an infinite-temperature state. The number of excited particles follows the Poisson distribution. The collision between particles is not affected by the random field. But the signals of colliding particles are gradually covered by the background excitations caused by random field.
Pei Wang, Yejie Wang, Muxi Diao, Keqing He, Guanting Dong, Weiran Xu
In the deployment of large language models (LLMs), accurate confidence estimation is critical for assessing the credibility of model predictions. However, existing methods often fail to overcome the issue of overconfidence on incorrect answers. In this work, we focus on improving the confidence estimation of large language models. Considering the fragility of self-awareness in language models, we introduce a Multi-Perspective Consistency (MPC) method. We leverage complementary insights from different perspectives within models (MPC-Internal) and across different models (MPC-Across) to mitigate the issue of overconfidence arising from a singular viewpoint. The experimental results on eight publicly available datasets show that our MPC achieves state-of-the-art performance. Further analyses indicate that MPC can mitigate the problem of overconfidence and is effectively scalable to other models.
Pei Wang, Gao Xianlong, Shaojun Xu
We study the transport through a quantum dot subject to a randomly fluctuating potential, generated by a sequence of pulses in the gate voltage with the help of the autoregressive model. We find that the tunneling current is multistable when the fluctuating potential with a finite correlation time is applied before the non-equilibrium steady state is built up. The non-equilibrium stationary current is heavily dependent on the history of the fluctuating potential during the transient period if the potential has a finite correlation time. Furthermore, the averaged current over the path of the fluctuating potential is a function of its strength and correlation time. Our work therefore provides a robust theoretical proposal for the controlling of the non-equilibrium stationary current through a quantum dot in a randomly fluctuating potential.
Pei Wang, Xuean Zhao, Ling Tang
We present the numerical operator method designed for the real time dynamics of currents through nanostructures beyond the linear response regime. We apply this method to the transient and stationary currents through nanostructures with different topologies, e.g., the flakes of square and honeycomb lattices. We find a quasi-stationary stage with a life proportional to the flake size in the transient currents through the square flakes, but this quasi-stationary stage is destroyed in the presence of disorder. However, there is no quasi-stationary stage in the transient currents through the honeycomb flakes, showing that the transient current depends strongly upon the topologies of the nanostructures. We also study the stationary current by taking the limit of the current at long times. We find that the stationary current through a square flake increases smoothly as the voltage bias increasing. In contrast, we find a threshold voltage in the current-voltage curve through a honeycomb flake, indicating a gap at the Fermi energy of a honeycomb flake.
Shuang Li, Li Hsu, Jie Peng, Pei Wang
Gaussian Graphical Models (GGMs) have been used to construct genetic regulatory networks where regularization techniques are widely used since the network inference usually falls into a high-dimension-low-sample-size scenario. Yet, finding the right amount of regularization can be challenging, especially in an unsupervised setting where traditional methods such as BIC or cross-validation often do not work well. In this paper, we propose a new method - Bootstrap Inference for Network COnstruction (BINCO) - to infer networks by directly controlling the false discovery rates (FDRs) of the selected edges. This method fits a mixture model for the distribution of edge selection frequencies to estimate the FDRs, where the selection frequencies are calculated via model aggregation. This method is applicable to a wide range of applications beyond network construction. When we applied our proposed method to building a gene regulatory network with microarray expression breast cancer data, we were able to identify high-confidence edges and well-connected hub genes that could potentially play important roles in understanding the underlying biological processes of breast cancer.
Rafael N. Alexander, Pei Wang, Niranjan Sridhar, Moran Chen, Olivier Pfister, Nicolas C. Menicucci
One-way quantum computing is experimentally appealing because it requires only local measurements on an entangled resource called a cluster state. Record-size, but non-universal, continuous-variable cluster states were recently demonstrated separately in the time and frequency domains. We propose to combine these approaches into a scalable architecture in which a single optical parametric oscillator and simple interferometer entangle up to ($3\times 10^3$ frequencies) $\times$ (unlimited number of temporal modes) into a new and computationally universal continuous-variable cluster state. We introduce a generalized measurement protocol to enable improved computational performance on this new entanglement resource.
Pei Wang
We propose random non-Hermitian Hamiltonians to model the generic stochastic nonlinear dynamics of a quantum state in Hilbert space. Our approach features an underlying linearity in the dynamical equations, ensuring the applicability of techniques used for solving linear systems. Additionally, it offers the advantage of easily incorporating statistical symmetry, a generalization of explicit symmetry to stochastic processes. To demonstrate the utility of our approach, we apply it to describe real-time dynamics, starting from an initial symmetry-preserving state and evolving into a randomly distributed, symmetry-breaking final state. Our model serves as a quantum framework for the transition process, from disordered states to ordered ones, where symmetry is spontaneously broken.
Pei Wang, Yanan Wu, Xiaoshuai Song, Weixun Wang, Gengru Chen, Zhongwen Li, Kezhong Yan, Ken Deng, Qi Liu, Shuaibing Zhao, Shaopan Xiong, Xuepeng Liu, Xuefeng Chen, Wanxi Deng, Wenbo Su, Bo Zheng
Large language model (LLM)-based agents are increasingly deployed in e-commerce shopping. To perform thorough, user-tailored product searches, agents should interpret personal preferences, engage in multi-turn dialogues, and ultimately retrieve and discriminate among highly similar products. However, existing research has yet to provide a unified simulation environment that consistently captures all of these aspects, and always focuses solely on evaluation benchmarks without training support. In this paper, we introduce ShopSimulator, a large-scale and challenging Chinese shopping environment. Leveraging ShopSimulator, we evaluate LLMs across diverse scenarios, finding that even the best-performing models achieve less than 40% full-success rate. Error analysis reveals that agents struggle with deep search and product selection in long trajectories, fail to balance the use of personalization cues, and to effectively engage with users. Further training exploration provides practical guidance for overcoming these weaknesses, with the combination of supervised fine-tuning (SFT) and reinforcement learning (RL) yielding significant performance improvements. Code and data will be released at https://github.com/ShopAgent-Team/ShopSimulator.
Pei Wang, Jian Li, Long Ji, Xian Hou, Erbil Gugercinoglu, Di Li, Diego F. Torres, Yutong Chen, Jiarui Niu, Weiwei Zhu, Bing Zhang, En-wei Liang, Li Zhang, Mingyu Ge, Zigao Dai, Lin Lin, Jinlin Han, Yi Feng, Chenhui Niu, Yongkun Zhang, Dengjiang Zhou, Heng Xu, Chunfeng Zhang, Jinchen Jiang, Chenchen Miao, Mao Yuan, Weiyang Wang, Youling Yue, Yunsheng Wu, Yabiao Wang, Chengjie Wang, Zhenye Gan, Yuxi Li, Zhongyi Sun, Mingmin Chi
Aug 17, 2023·astro-ph.HE·PDF Magnetars are neutron stars with extremely strong magnetic fields, frequently powering high-energy activity in X-rays. Pulsed radio emission following some X-ray outbursts have been detected (\citealt{Camilo2006,camilo2007a}), albeit its physical origin is unclear. It has long been speculated that the origin of magnetars' radio signals is different from those from canonical pulsars, although convincing evidence is still lacking. Five months after magnetar SGR 1935+2154's X-ray outburst and its associated Fast Radio Burst (FRB) 20200428, a radio pulsar phase was discovered. Here we report the discovery of X-ray spectral hardening associated with the emergence of periodic radio pulsations from SGR 1935+2154 and a detailed analysis of the properties of the radio pulses. The observations suggest that radio emission originates from the outer magnetosphere of the magnetar, and the surface heating due to the bombardment of inward-going particles from the radio emission region is responsible for the observed X-ray spectral hardening.
Xinxin Yang, Pei Wang
We study the transport through a resonant level coupled to two leads with the latter being described by Wigner's random matrices. By taking appropriate thermodynamic limit before taking the long time limit, we obtain the stationary current as a function of voltage bias. The I-V curve is similar to that of single impurity Anderson model. On the other hand, the current matrix and initial density matrix in our model look like random matrices in the eigenbasis of Hamiltonian. They satisfy the description of eigenstate thermalization hypothesis (ETH) and nonequilibrium steady state hypothesis (NESSH), respectively. A statistical formula of current has been derived based on ETH and NESSH (J. Stat. Mech.: Theo. Exp., 093105 (2017)). We check this formula in our model and find it to predict the stationary current to a high precision. The shape of I-V curve is explained by the peak structure in the characteristic function of NESSH, which is reminiscent of the transmission coefficient.
Jingqian Sun, Pei Wang, Ronghao Li, Mei Zhou
Tree skeleton plays an important role in tree structure analysis, forest inventory and ecosystem monitoring. However, it is a challenge to extract a skeleton from a tree point cloud with complex branches. In this paper, an automatic and fast tree skeleton extraction method (FTSEM) based on voxel thinning is proposed. In this method, a wood-leaf classification algorithm was introduced to filter leaf points for the reduction of the leaf interference on tree skeleton generation, tree voxel thinning was adopted to extract raw tree skeleton quickly, and a breakpoint connection algorithm was used to improve the skeleton connectivity and completeness. Experiments were carried out in Haidian Park, Beijing, in which 24 trees were scanned and processed to obtain tree skeletons. The graph search algorithm (GSA) is used to extract tree skeletons based on the same datasets. Compared with GSA method, the FTSEM method obtained more complete tree skeletons. And the time cost of the FTSEM method is evaluated using the runtime and time per million points (TPMP). The runtime of FTSEM is from 1.0 s to 13.0 s, and the runtime of GSA is from 6.4 s to 309.3 s. The average value of TPMP is 1.8 s for FTSEM, and 22.3 s for GSA respectively. The experimental results demonstrate that the proposed method is feasible, robust, and fast with a good potential on tree skeleton extraction.
Pei Wang, Yijun Li, Nuno Vasconcelos
Extensive research in neural style transfer methods has shown that the correlation between features extracted by a pre-trained VGG network has a remarkable ability to capture the visual style of an image. Surprisingly, however, this stylization quality is not robust and often degrades significantly when applied to features from more advanced and lightweight networks, such as those in the ResNet family. By performing extensive experiments with different network architectures, we find that residual connections, which represent the main architectural difference between VGG and ResNet, produce feature maps of small entropy, which are not suitable for style transfer. To improve the robustness of the ResNet architecture, we then propose a simple yet effective solution based on a softmax transformation of the feature activations that enhances their entropy. Experimental results demonstrate that this small magic can greatly improve the quality of stylization results, even for networks with random weights. This suggests that the architecture used for feature extraction is more important than the use of learned weights for the task of style transfer.
Pei Wang, Yijun Li, Krishna Kumar Singh, Jingwan Lu, Nuno Vasconcelos
We introduce an inversion based method, denoted as IMAge-Guided model INvErsion (IMAGINE), to generate high-quality and diverse images from only a single training sample. We leverage the knowledge of image semantics from a pre-trained classifier to achieve plausible generations via matching multi-level feature representations in the classifier, associated with adversarial training with an external discriminator. IMAGINE enables the synthesis procedure to simultaneously 1) enforce semantic specificity constraints during the synthesis, 2) produce realistic images without generator training, and 3) give users intuitive control over the generation process. With extensive experimental results, we demonstrate qualitatively and quantitatively that IMAGINE performs favorably against state-of-the-art GAN-based and inversion-based methods, across three different image domains (i.e., objects, scenes, and textures).
Honghao Yin, Shu Chen, Gao Xianlong, Pei Wang
We study the Loschmidt echo and the dynamical free energy of the Anderson model after a quench of the disorder strength. If the initial state is extended and the eigenstates of the post-quench Hamiltonian are strongly localized, we argue that the Loschmidt echo exhibits zeros periodically with the period $2π/D$ where $D$ is the width of spectra. At these zeros, the dynamical free energy diverges in a logarithmic way. We present numerical evidence of our argument in one- and three-dimensional Anderson models. Our findings connect the dynamical quantum phase transitions to the localization-delocalization phase transitions.
Pei Wang
In this paper we introduce a method of calculating the local temperature and chemical potential inside a mesoscopic device out of equilibrium. We show how to check the conditions of local thermal equilibrium as the whole system is out of equilibrium. Especially we study the onsite chemical potentials inside a chain coupled to two reservoirs at a finite voltage bias. In the presence of disorder we observe a large fluctuation in onsite chemical potentials, which can be suppressed by the electron-electron interaction. By taking average with respect to the configurations of disorder, we recover the classical picture where the voltage drops monotonously through the resistance wire. We prove the existence of local intensive variables in a mesoscopic device which is in equilibrium or not far from equilibrium.
Pei Wang, Keqing He, Yutao Mou, Xiaoshuai Song, Yanan Wu, Jingang Wang, Yunsen Xian, Xunliang Cai, Weiran Xu
Detecting out-of-domain (OOD) intents from user queries is essential for a task-oriented dialogue system. Previous OOD detection studies generally work on the assumption that plenty of labeled IND intents exist. In this paper, we focus on a more practical few-shot OOD setting where there are only a few labeled IND data and massive unlabeled mixed data that may belong to IND or OOD. The new scenario carries two key challenges: learning discriminative representations using limited IND data and leveraging unlabeled mixed data. Therefore, we propose an adaptive prototypical pseudo-labeling (APP) method for few-shot OOD detection, including a prototypical OOD detection framework (ProtoOOD) to facilitate low-resource OOD detection using limited IND data, and an adaptive pseudo-labeling method to produce high-quality pseudo OOD\&IND labels. Extensive experiments and analysis demonstrate the effectiveness of our method for few-shot OOD detection.
Pei Wang, Guochao Bu, Ronghao Li, Rui Zhao
Terrestrial laser scanner is a kind of fast, high-precision data acquisition device, which had been more and more applied to the research areas of forest inventory. In this study, a kind of automated low-cost terrestrial laser scanner was designed and implemented based on a two-dimensional laser radar sensor SICK LMS-511 and a stepper motor. The new scanner was named as BEE, which can scan the forest trees in three dimension. The BEE scanner and its supporting software are specifically designed for forest inventory. The experiments have been performed by using the BEE scanner in an artificial ginkgo forest which was located in Haidian district of Beijing. Four square plots were selected to do the experiments. The BEE scanner scanned in the four plots and acquired the single scan data respectively. The DBH, tree height and tree position of trees in the four plots were estimated and analyzed. For comparison, the manual measured data was also collected in the four plots. The tree stem detection rate for all four plots was 92.75%; the root mean square error of the DBH estimation was 1.27cm; the root mean square error of the tree height estimation was 0.24m; the tree position estimation was in line with the actual position. Experimental results show that the BEE scanner can efficiently estimate the structure parameters of forest trees and has a good potential in practical application of forest inventory.