BZPEERPapers, people, evidence and research paths
Menu

Discover

SearchFind papers, people, topics, institutions and opportunities from one place.GraphSee how papers, authors, topics and evidence connect.

Research tools

HomeContinue from the papers, people and topics that matter now.CollectionsTurn saved papers into reading lists, evidence maps and shareable context.ResearchReconstruct the path that led you from search to paper to question.CollaborationMove from a paper, topic or expert request into a focused thread.PublishCreate full-text research articles with metadata, formulas and versions.

Network

ExpertsFind specialists whose work and reviews explain why they can help.CommunitiesJoin research groups organized around papers, topics and shared evidence.

Opportunities

ListingsBrowse roles, calls and collaborations connected to real research context.

Account

Log inReturn to your saved papers, graph views and active threads.Create accountStart saving papers, building a research trail and joining teams.
Log inCreate account

Source author record

Huifang Xu

Huifang Xu appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record
physics.geo-phComputation and Languageeess.ASMachine Learning

Catalog footprint

What is connected

5works
4topics
4close collaborators

Actions

Connect this record

Log in to claim
Open graphBrowse works

Research graph

See the researcher in context

Open full explorer

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

5 published item(s)

preprint2020arXiv

Integrating Discrete and Neural Features via Mixed-feature Trans-dimensional Random Field Language Models

There has been a long recognition that discrete features (n-gram features) and neural network based features have complementary strengths for language models (LMs). Improved performance can be obtained by model interpolation, which is, however, a suboptimal two-step integration of discrete and neural features. The trans-dimensional random field (TRF) framework has the potential advantage of being able to flexibly integrate a richer set of features. However, either discrete or neural features are used alone in previous TRF LMs. This paper develops a mixed-feature TRF LM and demonstrates its advantage in integrating discrete and neural features. Various LMs are trained over PTB and Google one-billion-word datasets, and evaluated in N-best list rescoring experiments for speech recognition. Among all single LMs (i.e. without model interpolation), the mixed-feature TRF LMs perform the best, improving over both discrete TRF LMs and neural TRF LMs alone, and also being significantly better than LSTM LMs. Compared to interpolating two separately trained models with discrete and neural features respectively, the performance of mixed-feature TRF LMs matches the best interpolated model, and with simplified one-step training process and reduced training time.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Investigation of the role of polysaccharide in the dolomite growth at low temperature by using atomistic simulations

Dehydration of water from surface Mg2+ is most likely the rate-limiting step in the dolomite growth at low temperature. Here, we investigate the role of polysaccharide in this step using classical molecular dynamics (MD) calculations. Free energy (potential of mean force, PMF) calculations have been performed for water molecules leaving the first two hydration layers above the dolomite (104) surface under the following three conditions: without catalyst, with monosaccharide (mannose) and with oligosaccharide (three units of mannose). MD simulations reveal that there is no obvious effect of monosaccharide in lowering the dehydration barrier for surface Mg2+. However, we found that there are metastable configurations of oligosaccharide, which can decrease the dehydration barrier of surface Mg2+ by about 0.7-1.1 kcal/mol. In these configurations, the molecule lies relatively flat on the surface and forms a bridge shape. The hydrophobic space near the surface created by the non-polar -CH groups of the oligosaccharide in the bridge conformation is the reason for the observed reduction of dehydration barrier.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Modeling the Effect of Dissolved Hydrogen Sulfide on Mg2+-water Complex on Dolomite {104} Surfaces

The key kinetic barrier to dolomite formation is related to the surface Mg2+-H2O complex, which hinders binding of surface Mg2+ ions to the CO3 2- ions in solution. It has been proposed that this reaction can be catalyzed by dissolved hydrogen sulfide. To characterize the role of dissolved hydrogen sulfide in the dehydration of surface Mg 2+ ions, ab initio simulations based on density functional theory (DFT) were carried out to study the thermodynamics of competitive adsorption of hydrogen sulfide and water on dolomite (104) surfaces from solution. We find that water is thermodynamically more stable on the surface with the difference in adsorption energy of -13.6 kJ/mol (in vacuum) and -12.8 kJ/mol (in aqueous solution). However, aqueous hydrogen sulfide adsorbed on the surface increases the Mg2+-H2O distances on surrounding surface sites. Two possible mechanisms were proposed for the catalytic effects of adsorbed hydrogen sulfide on the anhydrous Ca-Mg-carbonate crystallization at low temperature.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

STEM investigation of exsolution lamellae and "c" reflections in Ca-rich dolomite from the Platteville Formation, western Wisconsin

Dolomite crystals in partially dolomitized limestone from the Platteville Formation are both compositionally and microstructurally heterogeneous. A single dolomite crystal usually contains three phases: the host Ca-rich dolomite [Ca1.14Mg0.86(CO3)2], an Fe-bearing dolomite [Ca1.06Mg0.80Fe0.14(CO3)2], and calcite inclusions. These three phases show similar orientations. The Ca-rich dolomite exhibits modulated microstructures with wavelength ranging from 7 to 30 nm. The modulated microstructures are not evident in Fe-bearing dolomite. Modulations in the Ca-rich dolomite have three predominant orientation ranges in the studied sample: from (205) to (104), from (001) to (-101), and ~(110), which are consistent with previous studies. Bright-field (BF) and high-angle annular dark-field (HAADF) images confirm that these modulations are due to chemical variation rather than strain or diffraction contrast. The Ca-rich lamellae are Mg-rich calcite with compositions ranging from Ca0.85Mg0.15CO3 to Ca0.70Mg0.30CO3. The observed results indicate that these Ca-rich exsolution lamellae formed during diagenesis. In this study, three kinds of "c"-reflections, which are weak spots in the halfway position between the principal reflections along the (104)*, (112)*, and (110)* directions, have been found in the diffraction patterns of some Ca-rich dolomite. Mg-Ca ordering in x-y planes was not observed directly in Z-contrast images. FFT patterns from the Z-contrast images do not show "c"-reflections. STEM images confirm that the "c"-reflections could result from multiple diffraction between the host dolomite and twinned Mg-calcite nano-lamellae under TEM imaging and diffraction modes.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Z-contrast imaging and ab initio study on "d" superstructure in sedimentary dolomite

Nano-precipitates with tripled periodicity along the c-axis are observed in a Ca-rich dolomite sample from Proterozoic carbonate rocks with "molar tooth" structure. This observation is consistent with previous description of d reflections. High-angle annular dark-field STEM imaging (or Z-contrast imaging) that avoids dynamic diffraction as seen in electron diffraction and high-resolution TEM imaging modes, confirms that d reflections correspond to nanoscale precipitates aligned parallel to (001) of the host dolomite. The lamellae precipitates have a cation ordering sequence of Ca-Ca-Mg-Ca-Ca- Mg along the c direction resulting in a chemical composition of Ca0.67Mg0.33CO3. This superstructure is attributed to the extra or d reflections, thus is referred to as the d superstructure in this study. The structure can be simply described as interstratified calcite/dolomite. The crystal structure of the d superstructure calculated from density functional theory (DFT) has a space group of P31c and has a and c unit-cell parameters of 4.879 and 16.260 Å, respectively, values between those of dolomite and calcite. The detailed structural characteristics and parameters obtained from ab initio calculations are also reported in this paper. The method of combining Z-contrast imaging and ab initio calculations can be used for solving structures of other nano-precipitates and nano-phases.

0 citations0 reviewsNo code linkedOpen access