arXiv subjects

Heng Ji

Publications and source records attributed to Heng Ji.

358 records · Page 20Linked to original sources

arXiv2017

Improving Slot Filling Performance with Attentive Neural Networks on Dependency Structures

Slot Filling (SF) aims to extract the values of certain types of attributes (or slots, such as person:cities\_of\_residence) for a given entity from a large collection of source documents. In this paper we propose an effective DNN architecture for SF with the following new strategies: (1). Take a regularized dependency graph instead of a raw sentence as input to DNN, to compress the wide contexts between query and candidate filler; (2). Incorporate two attention mechanisms: local attention learned from query and candidate filler, and global attention learned from external knowledge bases, to guide the model to better select indicative contexts to determine slot type. Experiments show that this framework outperforms state-of-the-art on both relation extraction (16\% absolute F-score gain) and slot filling validation for each individual system (up to 8.5\% absolute F-score gain).

cs.CL↗

arXiv2017

CoType: Joint Extraction of Typed Entities and Relations with Knowledge Bases

Extracting entities and relations for types of interest from text is important for understanding massive text corpora. Traditionally, systems of entity relation extraction have relied on human-annotated corpora for training and adopted an incremental pipeline. Such systems require additional human expertise to be ported to a new domain, and are vulnerable to errors cascading down the pipeline. In this paper, we investigate joint extraction of typed entities and relations with labeled data heuristically obtained from knowledge bases (i.e., distant supervision). As our algorithm for type labeling via distant supervision is context-agnostic, noisy training data poses unique challenges for the task. We propose a novel domain-independent framework, called CoType, that runs a data-driven text segmentation algorithm to extract entity mentions, and jointly embeds entity mentions, relation mentions, text features and type labels into two low-dimensional spaces (for entity and relation mentions respectively), where, in each space, objects whose types are close will also have similar representations. CoType, then using these learned embeddings, estimates the types of test (unlinkable) mentions. We formulate a joint optimization problem to learn embeddings from text corpora and knowledge bases, adopting a novel partial-label loss function for noisy labeled data and introducing an object "translation" function to capture the cross-constraints of entities and relations on each other. Experiments on three public datasets demonstrate the effectiveness of CoType across different domains (e.g., news, biomedical), with an average of 25% improvement in F1 score compared to the next best method.

cs.CL↗

arXiv2017

Consecutive Insulator-Metal-Insulator Phase Transitions of Vanadium Dioxide by Hydrogen Doping

We report modulation of a reversible phase transition in VO2 films by hydrogen doping. A metallic phase and a new insulating phase are successively observed at room temperature as the doping concentration increases. It is suggested that the polarized charges from doped hydrogens play an important role. These charges gradually occupy V3d-O2p hybridized orbitals and consequently modulate the filling of the VO2 crystal conduction band-edge states, which eventually evolve into new valence band-edge states. This demonstrates the exceptional sensitivity of VO2 electronic properties to electron concentration and orbital occupancy, providing key information for the phase transition mechanism.

cond-mat.str-el↗

arXiv2017

Mesoscopic quantum effects in a bad metal, hydrogen-doped vanadium dioxide

The standard treatment of quantum corrections to semiclassical electronic conduction assumes that charge carriers propagate many wavelengths between scattering events, and succeeds in explaining multiple phenomena (weak localization magnetoresistance (WLMR), universal conductance fluctuations, Aharonov-Bohm oscillations) observed in polycrystalline metals and doped semiconductors in various dimensionalities. We report apparent WLMR and conductance fluctuations in H$_{x}$VO$_{2}$, a poor metal (in violation of the Mott-Ioffe-Regel limit) stabilized by the suppression of the VO$_{2}$ metal-insulator transition through atomic hydrogen doping. Epitaxial thin films, single-crystal nanobeams, and nanosheets show similar phenomenology, though the details of the apparent WLMR seem to depend on the combined effects of the strain environment and presumed doping level. Self-consistent quantitative analysis of the WLMR is challenging given this and the high resistivity of the material, since the quantitative expressions for WLMR are derived assuming good metallicity. These observations raise the issue of how to assess and analyze mesoscopic quantum effects in poor metals.

cond-mat.mes-hall↗

arXiv2016

Aligning Coordinated Text Streams through Burst Information Network Construction and Decipherment

Aligning coordinated text streams from multiple sources and multiple languages has opened many new research venues on cross-lingual knowledge discovery. In this paper we aim to advance state-of-the-art by: (1). extending coarse-grained topic-level knowledge mining to fine-grained information units such as entities and events; (2). following a novel Data-to-Network-to-Knowledge (D2N2K) paradigm to construct and utilize network structures to capture and propagate reliable evidence. We introduce a novel Burst Information Network (BINet) representation that can display the most important information and illustrate the connections among bursty entities, events and keywords in the corpus. We propose an effective approach to construct and decipher BINets, incorporating novel criteria based on multi-dimensional clues from pronunciation, translation, burst, neighbor and graph topological structure. The experimental results on Chinese and English coordinated text streams show that our approach can accurately decipher the nodes with high confidence in the BINets and that the algorithm can be efficiently run in parallel, which makes it possible to apply it to huge amounts of streaming data for never-ending language and information decipherment.

cs.CL↗

arXiv2016

Building a Fine-Grained Entity Typing System Overnight for a New X (X = Language, Domain, Genre)

Recent research has shown great progress on fine-grained entity typing. Most existing methods require pre-defining a set of types and training a multi-class classifier from a large labeled data set based on multi-level linguistic features. They are thus limited to certain domains, genres and languages. In this paper, we propose a novel unsupervised entity typing framework by combining symbolic and distributional semantics. We start from learning general embeddings for each entity mention, compose the embeddings of specific contexts using linguistic structures, link the mention to knowledge bases and learn its related knowledge representations. Then we develop a novel joint hierarchical clustering and linking algorithm to type all mentions using these representations. This framework doesn't rely on any annotated data, predefined typing schema, or hand-crafted features, therefore it can be quickly adapted to a new domain, genre and language. Furthermore, it has great flexibility at incorporating linguistic structures (e.g., Abstract Meaning Representation (AMR), dependency relations) to improve specific context representation. Experiments on genres (news and discussion forum) show comparable performance with state-of-the-art supervised typing systems trained from a large amount of labeled data. Results on various languages (English, Chinese, Japanese, Hausa, and Yoruba) and domains (general and biomedical) demonstrate the portability of our framework.

cs.CL↗

arXiv2016

Label Noise Reduction in Entity Typing by Heterogeneous Partial-Label Embedding

Current systems of fine-grained entity typing use distant supervision in conjunction with existing knowledge bases to assign categories (type labels) to entity mentions. However, the type labels so obtained from knowledge bases are often noisy (i.e., incorrect for the entity mention's local context). We define a new task, Label Noise Reduction in Entity Typing (LNR), to be the automatic identification of correct type labels (type-paths) for training examples, given the set of candidate type labels obtained by distant supervision with a given type hierarchy. The unknown type labels for individual entity mentions and the semantic similarity between entity types pose unique challenges for solving the LNR task. We propose a general framework, called PLE, to jointly embed entity mentions, text features and entity types into the same low-dimensional space where, in that space, objects whose types are semantically close have similar representations. Then we estimate the type-path for each training example in a top-down manner using the learned embeddings. We formulate a global objective for learning the embeddings from text corpora and knowledge bases, which adopts a novel margin-based loss that is robust to noisy labels and faithfully models type correlation derived from knowledge bases. Our experiments on three public typing datasets demonstrate the effectiveness and robustness of PLE, with an average of 25% improvement in accuracy compared to next best method.

cs.CL↗

arXiv2015

A Novel Method of Encoded Multiplexing Readout for Micro-pattern Gas Detectors

The requirement of a large number of electronic channels poses a big challenge for Micro-pattern Gas Detector (MPGD) to achieve good spatial resolution. By using the redundancy that at least two neighboring strips record the signal of a particle, a novel method of encoded multiplexing readout for MPGDs is presented in this paper. The method offers a feasible and easily-extensible way of encoding and decoding, and can significantly reduce the number of readout channels. A verification test was carried out on a 5*5 cm2 Thick Gas Electron Multiplier (THGEM) detector using a 8 keV Cu X-ray source with 100um slit, where 166 strips are read out by 21 encoded readout channels. The test results show a good linearity in its position response, and the spatial resolution root-mean-square (RMS) of the test system is about 260 μm. This method has an attractive potential to build large area detectors and can be easily adapted to other detectors like MPGDs.

physics.ins-det↗

arXiv2015

A Dependency-Based Neural Network for Relation Classification

Previous research on relation classification has verified the effectiveness of using dependency shortest paths or subtrees. In this paper, we further explore how to make full use of the combination of these dependency information. We first propose a new structure, termed augmented dependency path (ADP), which is composed of the shortest dependency path between two entities and the subtrees attached to the shortest path. To exploit the semantic representation behind the ADP structure, we develop dependency-based neural networks (DepNN): a recursive neural network designed to model the subtrees, and a convolutional neural network to capture the most important features on the shortest path. Experiments on the SemEval-2010 dataset show that our proposed method achieves state-of-art results.

cs.CL↗

arXiv2015

Leveraging Deep Neural Networks and Knowledge Graphs for Entity Disambiguation

Entity Disambiguation aims to link mentions of ambiguous entities to a knowledge base (e.g., Wikipedia). Modeling topical coherence is crucial for this task based on the assumption that information from the same semantic context tends to belong to the same topic. This paper presents a novel deep semantic relatedness model (DSRM) based on deep neural networks (DNN) and semantic knowledge graphs (KGs) to measure entity semantic relatedness for topical coherence modeling. The DSRM is directly trained on large-scale KGs and it maps heterogeneous types of knowledge of an entity from KGs to numerical feature vectors in a latent space such that the distance between two semantically-related entities is minimized. Compared with the state-of-the-art relatedness approach proposed by (Milne and Witten, 2008a), the DSRM obtains 19.4% and 24.5% reductions in entity disambiguation errors on two publicly available datasets respectively.

cs.CL↗

arXiv2015

Very large magnetoresistance in Fe$_{0.28}$TaS$_{2}$ single crystals

Magnetic moments intercalated into layered transition metal dichalcogenides are an excellent system for investigating the rich physics associated with magnetic ordering in a strongly anisotropic, strong spin-orbit coupling environment. We examine electronic transport and magnetization in Fe$_{0.28}$TaS$_{2}$, a highly anisotropic ferromagnet with a Curie temperature $T_{\mathrm{C}} \sim 68.8~$K. We find anomalous Hall data confirming a dominance of spin-orbit coupling in the magnetotransport properties of this material, and a remarkably large field-perpendicular-to-plane MR exceeding 60% at 2 K, much larger than the typical MR for bulk metals, and comparable to state-of-the-art GMR in thin film heterostructures, and smaller only than CMR in Mn perovskites or high mobility semiconductors. Even within the Fe$_x$TaS$_2$ series, for the current $x$ = 0.28 single crystals the MR is nearly $100\times$ higher than that found previously in the commensurate compound Fe$_{0.25}$TaS$_{2}$. After considering alternatives, we argue that the large MR arises from spin disorder scattering in the strong spin-orbit coupling environment, and suggest that this can be a design principle for materials with large MR.

cond-mat.str-el↗

arXiv2014

Nanostructure Investigations of Nonlinear Differential Conductance in NdNiO$_3$ Thin Films

Transport measurements on thin films of NdNiO$_3$ reveal a crossover to a regime of pronounced nonlinear conduction below the well-known metal-insulator transition temperature. The evolution of the transport properties at temperatures well below this transition appears consistent with a gradual formation of a gap in the hole-like Fermi surface of this strongly correlated system. As $T$ is decreased below the nominal transition temperature, transport becomes increasily non-Ohmic, with a model of Landau-Zener breakdown becoming most suited for describing $I(V)$ characteristics as the temperature approaches 2~K.

cond-mat.str-el↗

arXiv2014

Thermally Driven Analog of the Barkhausen Effect at the Metal-Insulator Transition in Vanadium Dioxide

The physics of the metal-insulator transition (MIT) in vanadium dioxide remains a subject of intense interest. Because of the complicating effects of elastic strain on the phase transition, there is interest in comparatively strain-free means of examining VO2 material properties. We report contact-free, low-strain studies of the MIT through an inductive bridge approach sensitive to the magnetic response of VO2 powder. Rather than observing the expected step-like change in susceptibility at the transition, we argue that the measured response is dominated by an analog of the Barkhausen effect, due to the extremely sharp jump in the magnetic response of each grain as a function of time as the material is cycled across the phase boundary. This effect suggests that future measurements could access the dynamics of this and similar phase transitions.

cond-mat.str-el↗

arXiv2014

Hydrogen Diffusion and Stabilization in Single-crystal VO2 Micro/nanobeams by Direct Atomic Hydrogenation

We report measurements of the diffusion of atomic hydrogen in single crystalline VO2 micro/nanobeams by direct exposure to atomic hydrogen, without catalyst. The atomic hydrogen is generated by a hot filament, and the doping process takes place at moderate temperature (373 K). Undoped VO2 has a metal-to-insulator phase transition at ~340 K between a high-temperature, rutile, metallic phase and a low-temperature, monoclinic, insulating phase with a resistance exhibiting a semiconductor-like temperature dependence. Atomic hydrogenation results in stabilization of the metallic phase of VO2 micro/nanobeams down to 2 K, the lowest point we could reach in our measurement setup. Based on observing the movement of the hydrogen diffusion front in single crystalline VO2 beams, we estimate the diffusion constant for hydrogen along the c-axis of the rutile phase to be 6.7 x 10^{-10} cm^2/s at approximately 373 K, exceeding the value in isostructural TiO2 by ~ 38x. Moreover, we find that the diffusion constant along the c-axis of the rutile phase exceeds that along the equivalent a-axis of the monoclinic phase by at least three orders of magnitude. This remarkable change in kinetics must originate from the distortion of the "channels" when the unit cell doubles along this direction upon cooling into the monoclinic structure. Ab initio calculation results are in good agreement with the experimental trends in the relative kinetics of the two phases. This raises the possibility of a switchable membrane for hydrogen transport.

cond-mat.mtrl-sci↗

arXiv2014

In situ diffraction study of catalytic hydrogenation of VO2: Stable phases and origins of metallicity

Controlling electronic population through chemical doping is one way to tip the balance between competing phases in materials with strong electronic correlations. Vanadium dioxide exhibits a first-order phase transition at around 338 K between a high temperature, tetragonal, metallic state (T) and a low temperature, monoclinic, insulating state (M1), driven by electron-electron and electron-lattice interactions. Intercalation of VO2 with atomic hydrogen has been demonstrated, with evidence that this doping suppresses the transition. However, the detailed effects of intercalated H on the crystal and electronic structure of the resulting hydride have not been previously reported. Here we present synchrotron and neutron diffraction studies of this material system, mapping out the structural phase diagram as a function of temperature and hydrogen content. In addition to the original T and M1 phases, we find two orthorhombic phases, O1 and O2, which are stabilized at higher hydrogen content. We present density functional calculations that confirm the metallicity of these states and discuss the physical basis by which hydrogen stabilizes conducting phases, in the context of the metal-insulator transition.

cond-mat.mtrl-sci↗

arXiv2013

Tweet, but Verify: Epistemic Study of Information Verification on Twitter

While Twitter provides an unprecedented opportunity to learn about breaking news and current events as they happen, it often produces skepticism among users as not all the information is accurate but also hoaxes are sometimes spread. While avoiding the diffusion of hoaxes is a major concern during fast-paced events such as natural disasters, the study of how users trust and verify information from tweets in these contexts has received little attention so far. We survey users on credibility perceptions regarding witness pictures posted on Twitter related to Hurricane Sandy. By examining credibility perceptions on features suggested for information verification in the field of Epistemology, we evaluate their accuracy in determining whether pictures were real or fake compared to professional evaluations performed by experts. Our study unveils insight about tweet presentation, as well as features that users should look at when assessing the veracity of tweets in the context of fast-paced events. Some of our main findings include that while author details not readily available on Twitter feeds should be emphasized in order to facilitate verification of tweets, showing multiple tweets corroborating a fact misleads users to trusting what actually is a hoax. We contrast some of the behavioral patterns found on tweets with literature in Psychology research.

cs.SI↗