Searcharxiv⌕ Search

arXiv subjects

Haopeng Zhang

Publications and source records attributed to Haopeng Zhang.

54 records · Page 3Linked to original sources
arXiv2023

Unsupervised Multi-document Summarization with Holistic Inference

Multi-document summarization aims to obtain core information from a collection of documents written on the same topic. This paper proposes a new holistic framework for unsupervised multi-document extractive summarization. Our method incorporates the holistic beam search inference method associated with the holistic measurements, named Subset Representative Index (SRI). SRI balances the importance and diversity of a subset of sentences from the source documents and can be calculated in unsupervised and adaptive manners. To demonstrate the effectiveness of our method, we conduct extensive experiments on both small and large-scale multi-document summarization datasets under both unsupervised and adaptive settings. The proposed method outperforms strong baselines by a significant margin, as indicated by the resulting ROUGE scores and diversity measures. Our findings also suggest that diversity is essential for improving multi-document summary performance.

cs.CL↗
arXiv2023

Contrastive Hierarchical Discourse Graph for Scientific Document Summarization

The extended structural context has made scientific paper summarization a challenging task. This paper proposes CHANGES, a contrastive hierarchical graph neural network for extractive scientific paper summarization. CHANGES represents a scientific paper with a hierarchical discourse graph and learns effective sentence representations with dedicated designed hierarchical graph information aggregation. We also propose a graph contrastive learning module to learn global theme-aware sentence representations. Extensive experiments on the PubMed and arXiv benchmark datasets prove the effectiveness of CHANGES and the importance of capturing hierarchical structure information in modeling scientific papers.

cs.CL↗
arXiv2023

DiffuSum: Generation Enhanced Extractive Summarization with Diffusion

Extractive summarization aims to form a summary by directly extracting sentences from the source document. Existing works mostly formulate it as a sequence labeling problem by making individual sentence label predictions. This paper proposes DiffuSum, a novel paradigm for extractive summarization, by directly generating the desired summary sentence representations with diffusion models and extracting sentences based on sentence representation matching. In addition, DiffuSum jointly optimizes a contrastive sentence encoder with a matching loss for sentence representation alignment and a multi-class contrastive loss for representation diversity. Experimental results show that DiffuSum achieves the new state-of-the-art extractive results on CNN/DailyMail with ROUGE scores of $44.83/22.56/40.56$. Experiments on the other two datasets with different summary lengths also demonstrate the effectiveness of DiffuSum. The strong performance of our framework shows the great potential of adapting generative models for extractive summarization. To encourage more following work in the future, we have released our codes at \url{https://github.com/hpzhang94/DiffuSum}

cs.CL↗
arXiv2023

The role of radial migration on tracing lithium evolution in the Galactic disk

With the calculated guiding center radius $R_{guiding}$ and birth radius $R_{birth}$, we investigate the role of radial migration on the description of lithium evolution in the Galactic disk based on the upper envelope of the A(Li) vs. [Fe/H] diagram. Using migration distances, we find that stars in the solar neighborhood are born at different locations in the galactic disk, and cannot all be explained by models of chemical evolution in the solar neighborhood. It is found that the upper envelope of the A(Li) vs. [Fe/H] diagram varies significantly with $R_{birth}$, which explains the decrease of Li for super-metal-rich (SMR) stars because they are non-young stars born in the inner disk. The upper envelope of Li-$R_{birth}$ plane fits very well with chemical evolution models by Grisoni et al. for $R_{birth} = 7 - 12$ kpc, outside which young stars generally lack sufficient time to migrate to the solar neighborhood. For stars born in the solar neighborhood, the young open clusters and the upper envelope of field stars with age $<$ 3 Gyr fit well with theoretical prediction. We find that calculations using stars with ages less than 3 Gyr are necessary to obtain an undepleted Li upper envelope, and that stars with solar age (around 4.5 Gyr) have depleted around 0.3 dex from the original value based on the chemical evolution model of Grisoni et al.

astro-ph.GA↗
arXiv2022

A Rotation Meanout Network with Invariance for Dermoscopy Image Classification and Retrieval

The computer-aided diagnosis (CAD) system can provide a reference basis for the clinical diagnosis of skin diseases. Convolutional neural networks (CNNs) can not only extract visual elements such as colors and shapes but also semantic features. As such they have made great improvements in many tasks of dermoscopy images. The imaging of dermoscopy has no principal orientation, indicating that there are a large number of skin lesion rotations in the datasets. However, CNNs lack rotation invariance, which is bound to affect the robustness of CNNs against rotations. To tackle this issue, we propose a rotation meanout (RM) network to extract rotation-invariant features from dermoscopy images. In RM, each set of rotated feature maps corresponds to a set of outputs of the weight-sharing convolutions and they are fused using meanout strategy to obtain the final feature maps. Through theoretical derivation, the proposed RM network is rotation-equivariant and can extract rotation-invariant features when followed by the global average pooling (GAP) operation. The extracted rotation-invariant features can better represent the original data in classification and retrieval tasks for dermoscopy images. The RM is a general operation, which does not change the network structure or increase any parameter, and can be flexibly embedded in any part of CNNs. Extensive experiments are conducted on a dermoscopy image dataset. The results show our method outperforms other anti-rotation methods and achieves great improvements in dermoscopy image classification and retrieval tasks, indicating the potential of rotation invariance in the field of dermoscopy images.

cs.CV↗
arXiv2022

HEGEL: Hypergraph Transformer for Long Document Summarization

Extractive summarization for long documents is challenging due to the extended structured input context. The long-distance sentence dependency hinders cross-sentence relations modeling, the critical step of extractive summarization. This paper proposes HEGEL, a hypergraph neural network for long document summarization by capturing high-order cross-sentence relations. HEGEL updates and learns effective sentence representations with hypergraph transformer layers and fuses different types of sentence dependencies, including latent topics, keywords coreference, and section structure. We validate HEGEL by conducting extensive experiments on two benchmark datasets, and experimental results demonstrate the effectiveness and efficiency of HEGEL.

cs.CL↗
arXiv2022

Evidence for Co-rotation Origin of Super Metal Rich Stars in LAMOST-Gaia: Multiple Ridges with a Similar Slope in phi versus Lz Plane

Super metal-rich (SMR) stars in the solar neighborhood are thought to be born in the inner disk and came to present location by radial migration, which is most intense at the co-rotation resonance (CR) of the Galactic bar. In this work, we show evidence for the CR origin of SMR stars in LAMOST-Gaia by detecting six ridges and undulations in the phi versus Lz space coded by median VR, following a similar slope of -8 km/s kpc/deg. The slope is predicted by Monario et al.'s model for CR of a large and slow Galactic bar. For the first time, we show the variation of angular momentum with azimuths from -10 deg to 20 deg for two outer and broad undulations with negative VR around -18 km/s following this slope. The wave-like pattern with large amplitude outside CR and a wide peak of the second undulations indicate that minor merger of the Sagittarius dwarf galaxy with the disk might play a role besides the significant impact of CR of the Galactic bar.

astro-ph.GA↗
arXiv2022

Improving the Faithfulness of Abstractive Summarization via Entity Coverage Control

Abstractive summarization systems leveraging pre-training language models have achieved superior results on benchmark datasets. However, such models have been shown to be more prone to hallucinate facts that are unfaithful to the input context. In this paper, we propose a method to remedy entity-level extrinsic hallucinations with Entity Coverage Control (ECC). We first compute entity coverage precision and prepend the corresponding control code for each training example, which implicitly guides the model to recognize faithfulness contents in the training phase. We further extend our method via intermediate fine-tuning on large but noisy data extracted from Wikipedia to unlock zero-shot summarization. We show that the proposed method leads to more faithful and salient abstractive summarization in supervised fine-tuning and zero-shot settings according to our experimental results on three benchmark datasets XSum, Pubmed, and SAMSum of very different domains and styles.

cs.CL↗
arXiv2021

Radial Migration from Metallicity Gradient of Open Clusters and Outliers

Radial migration is an important process in the evolution of the Galactic disk. The metallicity gradient of open clusters and its outliers provide an effective way to probe for this process. In this work, we compile metallicity, age, and kinematic parameters for 225 open clusters and carry out a quantitative analysis of radial migration via the calculated migration distances. Based on clusters with age $< 0.5$ Gyr, we obtain the present-day metallicity gradient of $-0.074 \pm 0.007$ dex/kpc. Along this gradient distributes three sequences, and clusters in the upper, the middle, and the lower groups are found to be old outward-migrators, in-situ clusters, and inward-migrators, respectively. The migration distance increases with age, but its most effective time is probably less than 3 Gyr. The metallicity gradient breaks out at $R_g$ (guiding center radius) $\sim11.5$ kpc, which is caused by the lack of young open clusters in the outer disk and the presence of old outward-migrators in the upper sequence. It shows that this boundary is related to the different effects of radial migration between the inner and outer disks. We also found many special open clusters in and near the outer disk of $R > 11$ kpc and a steeper metallicity gradient from the inner disk of $R_g < 7$ kpc, which tells a complicated evolution history of the Galactic disk by different effects of stellar radial migration.

astro-ph.GA↗
arXiv2021

High-α-Metal-Rich stars in the LAMOST-MRS survey and its connection with the galactic bulge

We report the detection of a large sample of high-$α$-metal-rich stars on the low giant branch with $2.6 0.2$ than the comparison group. These properties strongly indicate its connection with the outer bar/bulge region at $R=3-5$ kpc. A tentative interpretation of this special group is that its stars were formed in the X-shaped bar/bulge region, close to its corotation radius, where radial migration is the most intense, and brings them to present locations at 9 kpc and beyond. Low eccentricities and slightly outward radial excursions of its stars are consistent with this scenario. Its kinematics (cold) and chemistry ($[α/Fe]$ $\sim 0.1$) further support the formation of the instability-driven X-shaped bar/bulge from the thin disk.

astro-ph.GA↗
arXiv2021

Centrality Meets Centroid: A Graph-based Approach for Unsupervised Document Summarization

Unsupervised document summarization has re-acquired lots of attention in recent years thanks to its simplicity and data independence. In this paper, we propose a graph-based unsupervised approach for extractive document summarization. Instead of ranking sentences by salience and extracting sentences one by one, our approach works at a summary-level by utilizing graph centrality and centroid. We first extract summary candidates as subgraphs based on centrality from the sentence graph and then select from the summary candidates by matching to the centroid. We perform extensive experiments on two bench-marked summarization datasets, and the results demonstrate the effectiveness of our model compared to state-of-the-art baselines.

cs.CL↗
arXiv2021

Histopathology WSI Encoding based on GCNs for Scalable and Efficient Retrieval of Diagnostically Relevant Regions

Content-based histopathological image retrieval (CBHIR) has become popular in recent years in the domain of histopathological image analysis. CBHIR systems provide auxiliary diagnosis information for pathologists by searching for and returning regions that are contently similar to the region of interest (ROI) from a pre-established database. While, it is challenging and yet significant in clinical applications to retrieve diagnostically relevant regions from a database that consists of histopathological whole slide images (WSIs) for a query ROI. In this paper, we propose a novel framework for regions retrieval from WSI-database based on hierarchical graph convolutional networks (GCNs) and Hash technique. Compared to the present CBHIR framework, the structural information of WSI is preserved through graph embedding of GCNs, which makes the retrieval framework more sensitive to regions that are similar in tissue distribution. Moreover, benefited from the hierarchical GCN structures, the proposed framework has good scalability for both the size and shape variation of ROIs. It allows the pathologist defining query regions using free curves according to the appearance of tissue. Thirdly, the retrieval is achieved based on Hash technique, which ensures the framework is efficient and thereby adequate for practical large-scale WSI-database. The proposed method was validated on two public datasets for histopathological WSI analysis and compared to the state-of-the-art methods. The proposed method achieved mean average precision above 0.857 on the ACDC-LungHP dataset and above 0.864 on the Camelyon16 dataset in the irregular region retrieval tasks, which are superior to the state-of-the-art methods. The average retrieval time from a database within 120 WSIs is 0.802 ms.

cs.CV↗
arXiv2020

Fully spin-polarized nodal chain state in half metal LiV$_2$O$_4$

Nodal-chain fermions, as novel topological states of matter, have been hotly discussed in non-magnetic materials. Here, by using first-principles calculations and symmetry analysis, we propose the realization of fully spin-polarized nodal chain in the half-metal state of LiV$_2$O$_4$ compound. The material naturally shows a ferromagnetic ground state, and takes on a half-metal band structure with only the bands from the spin-up channel present near the Fermi level. The spin-up bands cross with each other, which form two types of nodal loops. These nodal loops arise from band inversion and are under the protection of the glide mirror symmetries. Remarkably, we find the nodal loops conjunct with each other and form chain-like nodal structure. Correspondingly, the w-shaped surface states are also fully spin-polarized. The fully spin-polarized nodal chain identified here has not been proposed in realistic materials before. An effective model is constructed to describe the nature of nodal chain. The effects of the electron correlation, the lattice strains, and the spin-orbit coupling are discussed. The fully spin-polarized bulk nodal-chain and the associated nontrivial surface states for a half-metal may open novel applications in spintronics.

cond-mat.mtrl-sci↗
arXiv2020

Graph-Bert: Only Attention is Needed for Learning Graph Representations

The dominant graph neural networks (GNNs) over-rely on the graph links, several serious performance problems with which have been witnessed already, e.g., suspended animation problem and over-smoothing problem. What's more, the inherently inter-connected nature precludes parallelization within the graph, which becomes critical for large-sized graph, as memory constraints limit batching across the nodes. In this paper, we will introduce a new graph neural network, namely GRAPH-BERT (Graph based BERT), solely based on the attention mechanism without any graph convolution or aggregation operators. Instead of feeding GRAPH-BERT with the complete large input graph, we propose to train GRAPH-BERT with sampled linkless subgraphs within their local contexts. GRAPH-BERT can be learned effectively in a standalone mode. Meanwhile, a pre-trained GRAPH-BERT can also be transferred to other application tasks directly or with necessary fine-tuning if any supervised label information or certain application oriented objective is available. We have tested the effectiveness of GRAPH-BERT on several graph benchmark datasets. Based the pre-trained GRAPH-BERT with the node attribute reconstruction and structure recovery tasks, we further fine-tune GRAPH-BERT on node classification and graph clustering tasks specifically. The experimental results have demonstrated that GRAPH-BERT can out-perform the existing GNNs in both the learning effectiveness and efficiency.

cs.LG↗
arXiv2015

Factorization of View-Object Manifolds for Joint Object Recognition and Pose Estimation

Due to large variations in shape, appearance, and viewing conditions, object recognition is a key precursory challenge in the fields of object manipulation and robotic/AI visual reasoning in general. Recognizing object categories, particular instances of objects and viewpoints/poses of objects are three critical subproblems robots must solve in order to accurately grasp/manipulate objects and reason about their environments. Multi-view images of the same object lie on intrinsic low-dimensional manifolds in descriptor spaces (e.g. visual/depth descriptor spaces). These object manifolds share the same topology despite being geometrically different. Each object manifold can be represented as a deformed version of a unified manifold. The object manifolds can thus be parameterized by its homeomorphic mapping/reconstruction from the unified manifold. In this work, we develop a novel framework to jointly solve the three challenging recognition sub-problems, by explicitly modeling the deformations of object manifolds and factorizing it in a view-invariant space for recognition. We perform extensive experiments on several challenging datasets and achieve state-of-the-art results.

cs.CV↗
arXiv2014

Convergence Analysis and Parallel Computing Implementation for the Multiagent Coordination Optimization Algorithm

In this report, a novel variation of Particle Swarm Optimization (PSO) algorithm, called Multiagent Coordination Optimization (MCO), is implemented in a parallel computing way for practical use by introducing MATLAB built-in function "parfor" into MCO. Then we rigorously analyze the global convergence of MCO by means of semistability theory. Besides sharing global optimal solutions with the PSO algorithm, the MCO algorithm integrates cooperative swarm behavior of multiple agents into the update formula by sharing velocity and position information between neighbors to improve its performance. Numerical evaluation of the parallel MCO algorithm is provided in the report by running the proposed algorithm on supercomputers in the High Performance Computing Center at Texas Tech University. In particular, the optimal value and consuming time are compared with PSO and serial MCO by solving several benchmark functions in the literature, respectively. Based on the simulation results, the performance of the parallel MCO is not only superb compared with PSO for solving many nonlinear, noncovex optimization problems, but also is of high efficiency by saving the computational time.

math.OC↗
arXiv2014

Shape Primitive Histogram: A Novel Low-Level Face Representation for Face Recognition

We further exploit the representational power of Haar wavelet and present a novel low-level face representation named Shape Primitives Histogram (SPH) for face recognition. Since human faces exist abundant shape features, we address the face representation issue from the perspective of the shape feature extraction. In our approach, we divide faces into a number of tiny shape fragments and reduce these shape fragments to several uniform atomic shape patterns called Shape Primitives. A convolution with Haar Wavelet templates is applied to each shape fragment to identify its belonging shape primitive. After that, we do a histogram statistic of shape primitives in each spatial local image patch for incorporating the spatial information. Finally, each face is represented as a feature vector via concatenating all the local histograms of shape primitives. Four popular face databases, namely ORL, AR, Yale-B and LFW-a databases, are employed to evaluate SPH and experimentally study the choices of the parameters. Extensive experimental results demonstrate that the proposed approach outperform the state-of-the-arts.

cs.CV↗