arXiv subjects

Wei Lu

Publications and source records attributed to Wei Lu.

At least 325 records · Page 18Linked to original sources

arXiv2020

ENT-DESC: Entity Description Generation by Exploring Knowledge Graph

Previous works on knowledge-to-text generation take as input a few RDF triples or key-value pairs conveying the knowledge of some entities to generate a natural language description. Existing datasets, such as WIKIBIO, WebNLG, and E2E, basically have a good alignment between an input triple/pair set and its output text. However, in practice, the input knowledge could be more than enough, since the output description may only cover the most significant knowledge. In this paper, we introduce a large-scale and challenging dataset to facilitate the study of such a practical scenario in KG-to-text. Our dataset involves retrieving abundant knowledge of various types of main entities from a large knowledge graph (KG), which makes the current graph-to-sequence models severely suffer from the problems of information loss and parameter explosion while generating the descriptions. We address these challenges by proposing a multi-graph structure that is able to represent the original graph information more comprehensively. Furthermore, we also incorporate aggregation methods that learn to extract the rich graph information. Extensive experiments demonstrate the effectiveness of our model architecture.

cs.CL↗

arXiv2020

Reproducible and Interpretable Spiculation Quantification for Lung Cancer Screening

Spiculations are important predictors of lung cancer malignancy, which are spikes on the surface of the pulmonary nodules. In this study, we proposed an interpretable and parameter-free technique to quantify the spiculation using area distortion metric obtained by the conformal (angle-preserving) spherical parameterization. We exploit the insight that for an angle-preserved spherical mapping of a given nodule, the corresponding negative area distortion precisely characterizes the spiculations on that nodule. We introduced novel spiculation scores based on the area distortion metric and spiculation measures. We also semi-automatically segment lung nodule (for reproducibility) as well as vessel and wall attachment to differentiate the real spiculations from lobulation and attachment. A simple pathological malignancy prediction model is also introduced. We used the publicly-available LIDC-IDRI dataset pathologists (strong-label) and radiologists (weak-label) ratings to train and test radiomics models containing this feature, and then externally validate the models. We achieved AUC$=$0.80 and 0.76, respectively, with the models trained on the 811 weakly-labeled LIDC datasets and tested on the 72 strongly-labeled LIDC and 73 LUNGx datasets; the previous best model for LUNGx had AUC$=$0.68. The number-of-spiculations feature was found to be highly correlated (Spearman's rank correlation coefficient $ρ= 0.44$) with the radiologists' spiculation score. We developed a reproducible and interpretable, parameter-free technique for quantifying spiculations on nodules. The spiculation quantification measures was then applied to the radiomics framework for pathological malignancy prediction with reproducible semi-automatic segmentation of nodule. Using our interpretable features (size, attachment, spiculation, lobulation), we were able to achieve higher performance than previous models.

cs.CV↗

arXiv2020

Pre-training for Abstractive Document Summarization by Reinstating Source Text

Abstractive document summarization is usually modeled as a sequence-to-sequence (Seq2Seq) learning problem. Unfortunately, training large Seq2Seq based summarization models on limited supervised summarization data is challenging. This paper presents three pre-training objectives which allow us to pre-train a Seq2Seq based abstractive summarization model on unlabeled text. The main idea is that, given an input text artificially constructed from a document, a model is pre-trained to reinstate the original document. These objectives include sentence reordering, next sentence generation, and masked document generation, which have close relations with the abstractive document summarization task. Experiments on two benchmark summarization datasets (i.e., CNN/DailyMail and New York Times) show that all three objectives can improve performance upon baselines. Compared to models pre-trained on large-scale data (more than 160GB), our method, with only 19GB text for pre-training, achieves comparable results, which demonstrates its effectiveness.

cs.CL↗

arXiv2020

Lightweight, Dynamic Graph Convolutional Networks for AMR-to-Text Generation

AMR-to-text generation is used to transduce Abstract Meaning Representation structures (AMR) into text. A key challenge in this task is to efficiently learn effective graph representations. Previously, Graph Convolution Networks (GCNs) were used to encode input AMRs, however, vanilla GCNs are not able to capture non-local information and additionally, they follow a local (first-order) information aggregation scheme. To account for these issues, larger and deeper GCN models are required to capture more complex interactions. In this paper, we introduce a dynamic fusion mechanism, proposing Lightweight Dynamic Graph Convolutional Networks (LDGCNs) that capture richer non-local interactions by synthesizing higher order information from the input graphs. We further develop two novel parameter saving strategies based on the group graph convolutions and weight tied convolutions to reduce memory usage and model complexity. With the help of these strategies, we are able to train a model with fewer parameters while maintaining the model capacity. Experiments demonstrate that LDGCNs outperform state-of-the-art models on two benchmark datasets for AMR-to-text generation with significantly fewer parameters.

cs.CL↗

arXiv2020

Two are Better than One: Joint Entity and Relation Extraction with Table-Sequence Encoders

Named entity recognition and relation extraction are two important fundamental problems. Joint learning algorithms have been proposed to solve both tasks simultaneously, and many of them cast the joint task as a table-filling problem. However, they typically focused on learning a single encoder (usually learning representation in the form of a table) to capture information required for both tasks within the same space. We argue that it can be beneficial to design two distinct encoders to capture such two different types of information in the learning process. In this work, we propose the novel {\em table-sequence encoders} where two different encoders -- a table encoder and a sequence encoder are designed to help each other in the representation learning process. Our experiments confirm the advantages of having {\em two} encoders over {\em one} encoder. On several standard datasets, our model shows significant improvements over existing approaches.

cs.CL↗

arXiv2020

Attention Guided Graph Convolutional Networks for Relation Extraction

Dependency trees convey rich structural information that is proven useful for extracting relations among entities in text. However, how to effectively make use of relevant information while ignoring irrelevant information from the dependency trees remains a challenging research question. Existing approaches employing rule based hard-pruning strategies for selecting relevant partial dependency structures may not always yield optimal results. In this work, we propose Attention Guided Graph Convolutional Networks (AGGCNs), a novel model which directly takes full dependency trees as inputs. Our model can be understood as a soft-pruning approach that automatically learns how to selectively attend to the relevant sub-structures useful for the relation extraction task. Extensive results on various tasks including cross-sentence n-ary relation extraction and large-scale sentence-level relation extraction show that our model is able to better leverage the structural information of the full dependency trees, giving significantly better results than previous approaches.

cs.CL↗

arXiv2020

RPT: Learning Point Set Representation for Siamese Visual Tracking

While remarkable progress has been made in robust visual tracking, accurate target state estimation still remains a highly challenging problem. In this paper, we argue that this issue is closely related to the prevalent bounding box representation, which provides only a coarse spatial extent of object. Thus an effcient visual tracking framework is proposed to accurately estimate the target state with a finer representation as a set of representative points. The point set is trained to indicate the semantically and geometrically significant positions of target region, enabling more fine-grained localization and modeling of object appearance. We further propose a multi-level aggregation strategy to obtain detailed structure information by fusing hierarchical convolution layers. Extensive experiments on several challenging benchmarks including OTB2015, VOT2018, VOT2019 and GOT-10k demonstrate that our method achieves new state-of-the-art performance while running at over 20 FPS.

cs.CV↗

arXiv2020

Reasoning with Latent Structure Refinement for Document-Level Relation Extraction

Document-level relation extraction requires integrating information within and across multiple sentences of a document and capturing complex interactions between inter-sentence entities. However, effective aggregation of relevant information in the document remains a challenging research question. Existing approaches construct static document-level graphs based on syntactic trees, co-references or heuristics from the unstructured text to model the dependencies. Unlike previous methods that may not be able to capture rich non-local interactions for inference, we propose a novel model that empowers the relational reasoning across sentences by automatically inducing the latent document-level graph. We further develop a refinement strategy, which enables the model to incrementally aggregate relevant information for multi-hop reasoning. Specifically, our model achieves an F1 score of 59.05 on a large-scale document-level dataset (DocRED), significantly improving over the previous results, and also yields new state-of-the-art results on the CDR and GDA dataset. Furthermore, extensive analyses show that the model is able to discover more accurate inter-sentence relations.

cs.CL↗

arXiv2020

Learning to Recognize Discontiguous Entities

This paper focuses on the study of recognizing discontiguous entities. Motivated by a previous work, we propose to use a novel hypergraph representation to jointly encode discontiguous entities of unbounded length, which can overlap with one another. To compare with existing approaches, we first formally introduce the notion of model ambiguity, which defines the difficulty level of interpreting the outputs of a model, and then formally analyze the theoretical advantages of our model over previous existing approaches based on linear-chain CRFs. Our empirical results also show that our model is able to achieve significantly better results when evaluated on standard data with many discontiguous entities.

cs.CL↗

arXiv2020

High-throughput injection-acceleration of electron bunches from a linear accelerator to a laser wakefield accelerator

Plasma-based accelerators (PBAs) driven by either intense lasers (laser wakefield accelerators, LWFAs) or particle beams (plasma wakefield accelerators, PWFAs), can accelerate charged particles at extremely high gradients compared to conventional radio-frequency (RF) accelerators. In the past two decades, great strides have been made in this field, making PBA a candidate for next-generation light sources and colliders. However, these challenging applications necessarily require beams with good stability, high quality, controllable polarization and excellent reproducibility. To date, such beams are generated only by conventional RF accelerators. Therefore, it is important to demonstrate the injection and acceleration of beams first produced using a conventional RF accelerator, by a PBA. In some recent studies on LWFA staging and external injection-acceleration in PWFA only a very small fraction (from below 0.1% to few percent) of the injected charge (the coupling efficiency) was accelerated. For future colliders where beam energy will need to be boosted using multiple stages, the coupling efficiency per stage must approach 100%. Here we report the first demonstration of external injection from a photocathode-RF-gun-based conventional linear accelerator (LINAC) into a LWFA and subsequent acceleration without any significant loss of charge or degradation of quality, which is achieved by properly shaping and matching the beam into the plasma structure. This is an important step towards realizing a high-throughput, multi-stage, high-energy, hybrid conventional-plasma accelerator.

physics.plasm-ph↗

arXiv2020

Mining Commonsense Facts from the Physical World

Textual descriptions of the physical world implicitly mention commonsense facts, while the commonsense knowledge bases explicitly represent such facts as triples. Compared to dramatically increased text data, the coverage of existing knowledge bases is far away from completion. Most of the prior studies on populating knowledge bases mainly focus on Freebase. To automatically complete commonsense knowledge bases to improve their coverage is under-explored. In this paper, we propose a new task of mining commonsense facts from the raw text that describes the physical world. We build an effective new model that fuses information from both sequence text and existing knowledge base resource. Then we create two large annotated datasets each with approximate 200k instances for commonsense knowledge base completion. Empirical results demonstrate that our model significantly outperforms baselines.

cs.CL↗

arXiv2020

Region-of-interest micro-focus CT based on an all-optical inverse Compton scattering source

Micro-focus computed tomography (CT), enabling the reconstruction of hyperfine structure within objects, is a powerful nondestructive testing tool in many fields. Current X-ray sources for micro-focus CT are typically limited by their relatively low photon energy and low flux. An all-optical inverse Compton scattering source (AOCS) based on laser wakefield accelerator (LWFA) can generate intense quasi-monoenergetic X/gamma-ray pulses in the keV-MeV range with micron-level source size, and its potential application for micro-focus CT has become very attractive in recent years due to the fast pace progress made in LWFA. Here we report the first experimental demonstration of high-fidelity micro-focus CT using AOCS (~70 keV) by imaging and reconstructing a test object with complex inner structures. A region-of-interest (ROI) CT method is adopted to utilize the relatively small field-of-view (FOV) of AOCS to obtain high-resolution reconstruction. This demonstration of the ROI micro-focus CT based on AOCS is a key step for its application in the field of hyperfine nondestructive testing.

physics.app-ph↗

arXiv2020

Hidden information in fluctuation in small systems

The exploration of the rich dynamics of electrons is a frontier in fundamental nano-physics. The dynamical behavior of electrons is dominated by random and chaotic thermal motion with ultrafast ($\approx$ ps) and nanoscale scatterings. This generates fluctuating electromagnetic fields in close vicinity of the moving electrons. W studied this fluctuation in small structures and found that its spatial distribution is not uniform, the magnitude of the fluctuation depends on external parameters such as the size ( ~1 $μm$) and the shape of the structure and changes can occur by an order of magnitude. Our work opens the possibility of improving the signal to noise ratio in small devices and in manipulating microscopic electron kinematics through nano-optical techniques and to applications in thermal detectors and photothermal photovoltaics.

cond-mat.mes-hall↗

arXiv2020

Adversarial Deep Network Embedding for Cross-network Node Classification

In this paper, the task of cross-network node classification, which leverages the abundant labeled nodes from a source network to help classify unlabeled nodes in a target network, is studied. The existing domain adaptation algorithms generally fail to model the network structural information, and the current network embedding models mainly focus on single-network applications. Thus, both of them cannot be directly applied to solve the cross-network node classification problem. This motivates us to propose an adversarial cross-network deep network embedding (ACDNE) model to integrate adversarial domain adaptation with deep network embedding so as to learn network-invariant node representations that can also well preserve the network structural information. In ACDNE, the deep network embedding module utilizes two feature extractors to jointly preserve attributed affinity and topological proximities between nodes. In addition, a node classifier is incorporated to make node representations label-discriminative. Moreover, an adversarial domain adaptation technique is employed to make node representations network-invariant. Extensive experimental results demonstrate that the proposed ACDNE model achieves the state-of-the-art performance in cross-network node classification.

cs.SI↗

arXiv2020

Read Beyond the Lines: Understanding the Implied Textual Meaning via a Skim and Intensive Reading Model

The nonliteral interpretation of a text is hard to be understood by machine models due to its high context-sensitivity and heavy usage of figurative language. In this study, inspired by human reading comprehension, we propose a novel, simple, and effective deep neural framework, called Skim and Intensive Reading Model (SIRM), for figuring out implied textual meaning. The proposed SIRM consists of two main components, namely the skim reading component and intensive reading component. N-gram features are quickly extracted from the skim reading component, which is a combination of several convolutional neural networks, as skim (entire) information. An intensive reading component enables a hierarchical investigation for both local (sentence) and global (paragraph) representation, which encapsulates the current embedding and the contextual information with a dense connection. More specifically, the contextual information includes the near-neighbor information and the skim information mentioned above. Finally, besides the normal training loss function, we employ an adversarial loss function as a penalty over the skim reading component to eliminate noisy information arisen from special figurative words in the training data. To verify the effectiveness, robustness, and efficiency of the proposed architecture, we conduct extensive comparative experiments on several sarcasm benchmarks and an industrial spam dataset with metaphors. Experimental results indicate that (1) the proposed model, which benefits from context modeling and consideration of figurative language, outperforms existing state-of-the-art solutions, with comparable parameter scale and training speed; (2) the SIRM yields superior robustness in terms of parameter size sensitivity; (3) compared with ablation and addition variants of the SIRM, the final framework is efficient enough.

cs.CL↗

arXiv2019

Knowing What, How and Why: A Near Complete Solution for Aspect-based Sentiment Analysis

Target-based sentiment analysis or aspect-based sentiment analysis (ABSA) refers to addressing various sentiment analysis tasks at a fine-grained level, which includes but is not limited to aspect extraction, aspect sentiment classification, and opinion extraction. There exist many solvers of the above individual subtasks or a combination of two subtasks, and they can work together to tell a complete story, i.e. the discussed aspect, the sentiment on it, and the cause of the sentiment. However, no previous ABSA research tried to provide a complete solution in one shot. In this paper, we introduce a new subtask under ABSA, named aspect sentiment triplet extraction (ASTE). Particularly, a solver of this task needs to extract triplets (What, How, Why) from the inputs, which show WHAT the targeted aspects are, HOW their sentiment polarities are and WHY they have such polarities (i.e. opinion reasons). For instance, one triplet from "Waiters are very friendly and the pasta is simply average" could be ('Waiters', positive, 'friendly'). We propose a two-stage framework to address this task. The first stage predicts what, how and why in a unified model, and then the second stage pairs up the predicted what (how) and why from the first stage to output triplets. In the experiments, our framework has set a benchmark performance in this novel triplet extraction task. Meanwhile, it outperforms a few strong baselines adapted from state-of-the-art related methods.

cs.CL↗

arXiv2019

The Parameters of Minimal Linear Codes

Let $k\leq n$ be two positive integers and $q$ a prime power. The basic question in minimal linear codes is to determine if there exists an $[n,k]_q$ minimal linear code. The first objective of this paper is to present a new sufficient and necessary condition for linear codes to be minimal. Using this condition, it is easy to construct minimal linear codes or to prove some linear codes are minimal. The second objective of this paper is to use the new sufficient and necessary condition to partially give an answer to the basic question in minimal linear codes. The third objective of this paper is to present four classes of minimal linear codes, which generalize the results about the binary case given in [X. Li and Q. Yue, Des. Codes Cryptogr., 2019]. One can find that our method is much easier and more effective.

cs.IT↗

arXiv2019

Six constructions of asymptotically optimal codebooks via the character sums

In this paper, using additive characters of finite field, we find a codebook which is equivalent to the measurement matrix in [20]. The advantage of our construction is that it can be generalized naturally to construct the other five classes of codebooks using additive and multiplicative characters of finite field. We determine the maximal cross-correlation amplitude of these codebooks by the properties of characters and character sums. We prove that all the codebooks we constructed are asymptotically optimal with respect to the Welch bound. The parameters of these codebooks are new.

cs.IT↗