In the Chinese medical insurance industry, the assessors role is essential and requires significant efforts to converse with the claimant. This is a highly professional job that involves many parts, such as identifying personal information, collectin
g related evidence, and making a final insurance report. Due to the coronavirus (COVID-19) pandemic, the previous offline insurance assessment has to be conducted online. However, for the junior assessor often lacking practical experience, it is not easy to quickly handle such a complex online procedure, yet this is important as the insurance company needs to decide how much compensation the claimant should receive based on the assessors feedback. In order to promote assessors work efficiency and speed up the overall procedure, in this paper, we propose a dialogue-based information extraction system that integrates advanced NLP technologies for medical insurance assessment. With the assistance of our system, the average time cost of the procedure is reduced from 55 minutes to 35 minutes, and the total human resources cost is saved 30% compared with the previous offline procedure. Until now, the system has already served thousands of online claim cases.
Recent advances in topological mechanics have revealed unusual phenomena such as topologically protected floppy modes and states of self-stress that are exponentially localized at boundaries and interfaces of mechanical networks. In this paper, we ex
plore the topological mechanics of epithelial tissues, where the appearance of these boundary and interface modes could lead to localized soft or stressed spots and play a role in morphogenesis. We consider both a simple vertex model (VM) governed by an effective elastic energy and its generalization to an active tension network (ATN) which incorporates active adaptation of the cytoskeleton. By analyzing spatially periodic lattices at the Maxwell point of mechanical instability, we find topologically polarized phases with exponential localization of floppy modes and states of self-stress in the ATN when cells are allowed to become concave, but not in the VM.
Despite the extensive studies of topological states, their characterization in strongly nonlinear classical systems has been lacking. In this work, we identify the proper definition of Berry phase for nonlinear bulk modes and characterize topological
phases in one-dimensional (1D) generalized nonlinear Schr{o}dinger equations in the strongly nonlinear regime. We develop an analytic strategy to demonstrate the quantization of nonlinear Berry phase due to reflection symmetry. Mode amplitude itself plays a key role in nonlinear modes and controls topological phase transitions. We then show bulk-boundary correspondence by identifying the associated nonlinear topological edge modes. Interestingly, anomalous topological modes decay away from lattice boundaries to plateaus governed by fixed points of nonlinearities. We propose passive photonic and active electrical systems that can be experimentally implemented. Our work opens the door to the rich physics between topological phases of matter and nonlinear dynamics.
We conducted a joint experimental-theoretical investigation of the high-pressure chemistry of europium polyhydrides at pressures of 86-130 GPa. We discovered several novel magnetic Eu superhydrides stabilized by anharmonic effects: cubic $EuH_{9}$, h
exagonal $EuH_{9}$, and an unexpected cubic (Pm-3n) clathrate phase, $Eu_{8}H_{46}$. Monte Carlo simulations indicate that cubic $EuH_{9}$ has antiferromagnetic ordering with T(Neel) up to 24 K, whereas hexagonal $EuH_{9}$ and Pm-3n-$Eu_{8}H_{46}$ possess ferromagnetic ordering with T(Curie) = 137 and 336 K, respectively. The electron-phonon interaction is weak in all studied europium hydrides, and their magnetic ordering excludes s-wave superconductivity, except, perhaps, for distorted pseudohexagonal $EuH_{9}$. The equations of state predicted within the DFT+U approach (the Hubbard corrections were found within linear response theory) are in close agreement with the experimental data. This work shows the great influence of the atomic radius on symmetry-breaking distortions of the crystal structures of superhydrides and on their thermodynamic stability.
Entities may have complex interactions in a knowledge graph (KG), such as multi-step relationships, which can be viewed as graph contextual information of the entities. Traditional knowledge representation learning (KRL) methods usually treat a singl
e triple as a training unit, and neglect most of the graph contextual information exists in the topological structure of KGs. In this study, we propose a Path-based Pre-training model to learn Knowledge Embeddings, called PPKE, which aims to integrate more graph contextual information between entities into the KRL model. Experiments demonstrate that our model achieves state-of-the-art results on several benchmark datasets for link prediction and relation prediction tasks, indicating that our model provides a feasible way to take advantage of graph contextual information in KGs.
Natural language data exhibit tree-like hierarchical structures such as the hypernym-hyponym relations in WordNet. FastText, as the state-of-the-art text classifier based on shallow neural network in Euclidean space, may not model such hierarchies pr
ecisely with limited representation capacity. Considering that hyperbolic space is naturally suitable for modeling tree-like hierarchical data, we propose a new model named HyperText for efficient text classification by endowing FastText with hyperbolic geometry. Empirically, we show that HyperText outperforms FastText on a range of text classification tasks with much reduced parameters.
Coprime arrays enable Direction-of-Arrival (DoA) estimation of an increased number of sources. To that end, the receiver estimates the autocorrelation matrix of a larger virtual uniform linear array (coarray), by applying selection or averaging to th
e physical arrays autocorrelation estimates, followed by spatial-smoothing. Both selection and averaging have been designed under no optimality criterion and attain arbitrary (suboptimal) Mean-Squared-Error (MSE) estimation performance. In this work, we design a novel coprime array receiver that estimates the coarray autocorrelations with Minimum-MSE (MMSE), for any probability distribution of the source DoAs. Our extensive numerical evaluation illustrates that the proposed MMSE approach returns superior autocorrelation estimates which, in turn, enable higher DoA estimation performance compared to standard counterparts.
We establish non-Hermitian topological mechanics in one dimensional (1D) and two dimensional (2D) lattices consisting of mass points connected by meta-beams that lead to odd elasticity. Extended from the non-Hermitian skin effect in 1D systems, we de
monstrate this effect in 2D lattices in which bulk elastic waves exponentially localize in both lattice directions. We clarify a proper definition of Berry phase in non-Hermitian systems, with which we characterize the lattice topology and show the emergence of topological modes on lattice boundaries. The eigenfrequencies of topological modes are complex due to the breaking of $mathcal{PT}$-symmetry and the excitations could exponentially grow in time in the damped regime. Besides the bulk modes, additional localized modes arise in the bulk band and they are easily affected by perturbations. These distinguishing features may manifest themselves in various active materials and biological systems.
Complex node interactions are common in knowledge graphs, and these interactions also contain rich knowledge information. However, traditional methods usually treat a triple as a training unit during the knowledge representation learning (KRL) proced
ure, neglecting contextualized information of the nodes in knowledge graphs (KGs). We generalize the modeling object to a very general form, which theoretically supports any subgraph extracted from the knowledge graph, and these subgraphs are fed into a novel transformer-based model to learn the knowledge embeddings. To broaden usage scenarios of knowledge, pre-trained language models are utilized to build a model that incorporates the learned knowledge representations. Experimental results demonstrate that our model achieves the state-of-the-art performance on several medical NLP tasks, and improvement above TransE indicates that our KRL method captures the graph contextualized information effectively.
The current search for room-temperature superconductivity is inspired by the unique properties of the electron-phonon interaction in metal superhydrides. Encouraged by the recently found highest-$T_C$ superconductor fcc-$LaH_{10}$, here we discover s
everal superhydrides of another lanthanide - neodymium. We identify three novel metallic Nd-H phases at pressure range from 85 to 135 GPa: $I4/mmm$-$NdH_4$, $C2/c$-$NdH_7$, $P6_3/mmc$-$NdH_9$, synthesized by laser-heating metal samples in NH3BH3 media for in situ generation of hydrogen. A lower trihydride $Fmbar{3}m$-$NdH_3$ is found at pressures from 2 to 52 GPa. $I4/mmm$-$NdH_4$ and $C2/c$-$NdH_7$ are stable from 135 down to 85 GPa, and $P6_3/mmc$-$NdH_9$ from 110 to 130 GPa. Measurements of the electrical resistance of NdH9 demonstrate a possible superconducting transition at ~4.5 K in $P6_3/mmc$-$NdH_9$. Our theoretical calculations predict that all the neodymium hydrides have antiferromagnetic order at pressures below 150 GPa and represent one of the first discovered examples of strongly correlated superhydrides with large exchange spin-splitting in the electron band structure (> 450 meV). The critical N$e$el temperatures for new neodymium hydrides are estimated using the mean-field approximation as about 4 K ($NdH_4$), 251 K ($NdH_7$) and 136 K ($NdH_9$).
