Search

Gallenene is a promising low-dimensional material with a structure down to the thickness of a single atom, similar to graphene. However, van der Waals stacking of two-dimensional (2D) gallenene under confinement remain poorly understood. In this study, we present evidence of the formation of parallel-stacked hexagonal gallenene (a100) structures in liquid gallium. The present study demonstrates the AB stacking of 2D gallenene a100 crystals in liquid gallium sandwiched between two graphene layers, as observed through transmission electron microscopy. A nonlinear optical response of the confined hexagonal gallenene was investigated through second harmonic generation (SHG) microscopy. The SHG signal exhibits periodic peak intensity shifts upon angular rotation up to 90 degrees and intensity dampening at elevated temperatures. These findings offer insights on device applications of 2D gallenene., Comment: 14 pages and 5 figures

We propose differential phase contrast (DPC) imaging using energy-filtered electrons to image the magnetic properties of materials at the atomic scale. Compared to DPC measurements with elastic electrons, our simulations predict about two orders of magnitude higher relative magnetic signal intensities and sensitivity to all three vector components of magnetization., Comment: 6 pages, 5 figures

Anion doping offers immense potential for tailoring material properties, but achieving precise control over anion incorporation remains a challenge due to complex synthesis processes and limitations in local dopant detection. Here, we investigate the F-ion intercalation within an infinite layer NdNiO2+x/SrTiO3 superlattice film using a two-step synthesis approach. We employ advanced four-dimensional scanning transmission electron microscopy (4D-STEM) coupled with electron energy loss spectroscopy to map the F distribution and its impact on the atomic and electronic structure. Our observations reveal a striking geometric reconstruction of the infinite layer structure upon fluorination, resulting in a more distorted orthorhombic phase compared to the pristine perovskite. Notably, F-ion intercalation occurs primarily at the apical sites of the polyhedron, with some occupation of basal sites in localized regions. This process leads to the formation of two distinct domains within the nickelate layer, reflecting a competition between polyhedral distortion and geometric frustration-induced neodymium (Nd) displacement near domain interfaces. Interestingly, we observe an anomalous structural distortion where basal site anions are displaced in the same direction as Nd atoms, potentially linked to the partial basal site F-ion occupation. This coexistence of diverse structural distortions signifies a locally disordered F-ion distribution within the infinite layer structure with distinct F-ion configurations. These findings provide crucial insights into understanding and manipulating anion doping at the atomic level, paving the way for the development of novel materials with precisely controlled functionalities.

This tutorial explains how to use piecewise cubic B\'ezier curves to draw arbitrarily oriented ellipses and elliptical arcs. The geometric principles discussed here result in strikingly simple interfaces for graphics functions that can draw (approximate) circles, ellipses, and arcs of circles and ellipses. C++ source code listings are included for these functions. Their code size can be relatively small because they are designed to be used with a graphics library or platform that draws B\'ezier curves, and the library or platform is tasked with the actual rendering of the curves., Comment: 23 pages, 8 figures

Sapphire is a technologically highly relevant material, but it poses many challenges to performing epitaxial thin-film deposition. We have identified and applied the conditions for adsorption-controlled homoepitaxial growth of c-plane sapphire. The films thus grown are atomically smooth, have a controlled termination, and are of outstanding crystallinity. Their chemical purity exceeds that of the substrates. The films exhibit exceptional optical properties such as a single-crystal-like bandgap and a low density of F+ centers.

Superconducting films of alpha-Ta are promising candidates for the fabrication of advanced superconducting qubits. However, alpha-Ta films suffer from many growth-induced structural inadequacies that negatively affect their performance. We have therefore explored a new synthesis method for alpha-Ta films which allows the growth of these films with unprecedented quality. Using this method, high quality alpha-Ta films are deposited at a comparably high substrate temperature of 1150 C. They are single-phase alpha-Ta and have a single out-of-plane (110) orientation. They consist of grains above 2 um that have one of three possible in-plane orientations. As shown by scanning transmission electron microscopy and electron energy loss studies the substrate-film interfaces are sharp with no observable intermixing. The obtained insights into the epitaxial growth of body-centered-cubic films on quasi-hexagonal substrates lay the basis for harnessing the high structural coherence of such films in various applications.

Infinite-layer nickelates stand as a promising frontier in the exploration of unconventional superconductivity. Their synthesis through topotactic oxygen reduction from the parent perovskite phase remains a complex and elusive process. This study delves into the nano-scale effects of the topotactic lattice transformation within LaNiO$_2$ crystals. Leveraging high-resolution scanning transmission electron microscopy and spectroscopy, our investigations uncover a panorama of structural alterations, including grain boundaries and coherent twin boundaries, triggered by reduction-induced transformations. In addition, our analyses unveil the formation of an oxygen-rich disordered transition phase encircling impurities and pervading crystalline domains, and the internal strain is accommodated by grain boundary formation. By unraveling these nano-scale effects, our findings provide insights into the microscopic intricacies of the topotactic reduction process elucidating the transition from the perovskite to the infinite-layer phase within nickelate bulk crystals., Comment: 7 pages, 3 figures

Defect centers in hexagonal boron nitride have been extensively studied as room temperature single photon sources. The electronic structure of these defects exhibits strong coupling to phonons, as evidenced by the observation of phonon sidebands in both photoluminescence and cathodoluminescence spectra. However, the dynamics of the electron phonon coupling as well as phonon mediated dephasing of the color centers in hexagonal boron nitride remain unexplored. Here, we apply a novel time resolved CL spectroscopy technique to explore the population decay to phonon states and the dephasing time T2 with sub femtosecond time resolution. We demonstrate an ultrafast dephasing time of only 200 fs and a radiative decay of about 585 fs at room temperature, in contrast with all optical time resolved photoluminescence techniques that report a decay of a few nanoseconds. This behavior is attributed to efficient electron-beam excitation of coherent phonon polaritons in hexagonal boron nitride, resulting in faster dephasing of electronic transitions. Our results demonstrate the capability of our sequential cathodoluminescence spectroscopy technique to probe the ultrafast dephasing time of single emitters in quantum materials with sub femtosecond time resolution, heralding access to quantum path interferences in single emitters coupled to their complex environment.

Epitaxial growth of thin-film heterostructures is generally considered the most successful procedure to obtain interfaces of excellent structural and electronic quality between three-dimensional materials. However, these interfaces can only join material systems with crystal lattices of matching symmetries and lattice constants. We present a novel category of interfaces, the fabrication of which is membrane-based and does not require epitaxial growth. These interfaces therefore overcome limitations imposed by epitaxy. Leveraging the additional degrees of freedom gained, we demonstrate atomically clean interfaces between three-fold symmetric sapphire and four-fold symmetric SrTiO3. Atomic-resolution imaging reveals structurally well-defined interfaces with a novel moir\'e-type reconstruction.

Recent x-ray spectroscopic studies have revealed spin excitations and charge density waves in thin films of infinite-layer (IL) nickelates. However, clarifying whether the origin of these phenomena is intrinsic to the material class or attributable to impurity phases in the films has presented a major challenge. Here we utilize topotactic methods to synthesize bulk crystals of the IL nickelate LaNiO$_2$ with crystallographically oriented surfaces. We examine these crystals using resonant inelastic x-ray scattering (RIXS) at the Ni $L_3$-edge to elucidate the spin and charge correlations in the bulk of the material. While we detect the presence of prominent spin excitations in the crystals, fingerprints of charge order are absent at the ordering vectors identified in previous in thin-film studies. These results contribute to the understanding of the bulk properties of LaNiO$_2$ and establish topotactically synthesized crystals as viable complementary specimens for spectroscopic investigations., Comment: 11 pages, 7 figures with supplemental materials

The discovery of high-temperature superconductivity in La$_3$Ni$_2$O$_7$ at pressures above 14 GPa has spurred extensive research efforts. Yet, fundamental aspects of the superconducting phase, including the possibility of a filamentary character, are currently subjects of controversial debates. Conversely, a crystal structure with NiO$_6$ octahedral bilayers stacked along the $c$-axis direction was consistently posited in initial studies on La$_3$Ni$_2$O$_7$. Here we reassess this structure in optical floating zone-grown La$_3$Ni$_2$O$_7$ single crystals that show signs of filamentary superconductivity. Employing scanning transmission electron microscopy and single-crystal x-ray diffraction under high pressures, we observe multiple crystallographic phases in these crystals, with the majority phase exhibiting alternating monolayers and trilayers of NiO$_6$ octahedra, signifying a profound deviation from the previously suggested bilayer structure. Using density functional theory, we disentangle the individual contributions of the monolayer and trilayer structural units to the electronic band structure of La$_3$Ni$_2$O$_7$, providing a firm basis for advanced theoretical modeling and future evaluations of the potential of the monolayer-trilayer structure for hosting superconductivity.

For epitaxial Sr$_3$Ru$_2$O$_7$ films grown by pulsed laser deposition, we report a combined structural and magnetotransport study of thin films and Hall bar devices patterned side-by-side on the same film. Structural properties of these films are investigated using X-ray diffraction and high-resolution transmission electron microscopy, and confirm that these films are epitaxially oriented and nearly phase pure. For magnetic fields applied along the $c-$axis, a positive magnetoresistance of 10\% is measured for unpatterned Sr$_3$Ru$_2$O$_7$ films, whereas for patterned Hall bar devices of channel widths of $10$ and $5\, \mu$m, magnetoresistance values of 40\% and 140\% are found, respectively. These films show switching behaviors from positive to negative magnetoresistance that are controlled by the direction of the applied magnetic field. The present results provide a promising route for achieving stable epitaxial synthesis of intermediate members of correlated layered strontium ruthenates, and for the exploration of device physics in thin films of these compounds., Comment: 9 pages and 6 Figures in the Main text. 3 pages and 4 Figures in the Supplementary Material

Many recent improvements in NLP stem from the development and use of large pre-trained language models (PLMs) with billions of parameters. Large model sizes makes computational cost one of the main limiting factors for training and evaluating such models; and has raised severe concerns about the sustainability, reproducibility, and inclusiveness for researching PLMs. These concerns are often based on personal experiences and observations. However, there had not been any large-scale surveys that investigate them. In this work, we provide a first attempt to quantify these concerns regarding three topics, namely, environmental impact, equity, and impact on peer reviewing. By conducting a survey with 312 participants from the NLP community, we capture existing (dis)parities between different and within groups with respect to seniority, academia, and industry; and their impact on the peer reviewing process. For each topic, we provide an analysis and devise recommendations to mitigate found disparities, some of which already successfully implemented. Finally, we discuss additional concerns raised by many participants in free-text responses.

The strong structure-property coupling in rare-earth nickelates has spurred the realization of new quantum phases in rapid succession. Recently, topotactic transformations have provided a new platform for the controlled creation of oxygen vacancies and, therewith, for the exploitation of such coupling in nickelates. Here, we report the emergence of oxygen vacancy ordering in Pr$_{0.92}$Ca$_{0.08}$NiO$_{2.75}$ single crystals obtained via a topotactic reduction of the perovskite phase Pr$_{0.92}$Ca$_{0.08}$NiO$_{3}$, using CaH$_2$ as the reducing agent. We unveil a brownmillerite-like ordering pattern of the vacancies by high-resolution scanning transmission electron microscopy, with Ni ions in alternating square-pyramidal and octahedral coordination along the pseudocubic [100] direction. Furthermore, we find that the crystal structure acquires a high level of internal strain, where wavelike modulations of polyhedral tilts and rotations accommodate the large distortions around the vacancy sites. Our results suggest that atomic-resolution electron microscopy is a powerful method to locally resolve unconventional crystal structures that result from the topotactic transformation of complex oxide materials., Comment: 9 pages, 4 figures

We carefully investigated the storage of lithium in titania films on various substrates as a function of thickness. The experiments enable us to precisely separate contributions from bulk and boundary storage. The battery capacity measurements are complemented by bias dependent measurements of impedance, yielding interfacial resistance as well as interfacial capacitance. Independent information on electron and Li distribution is gained by scanning transmission electron microscopy (STEM), electron energy loss spectroscopy (EELS), aberration-corrected annular-bright-field (ABF) STEM. As a result, we obtain the full picture in terms of equilibrium storage (lithium content) and charge carrier concentrations as a function of spatial coordinates with cell voltage as a parameter. More importantly, both bulk storage which obeys electroneutrality and boundary storage which follows the space charge picture can be traced back to a common thermodynamic conception, and are obtained from it as special cases. This corresponds to no less than a unification of intercalation storage and super-capacitive storage, which are usually considered as independent phenomena, the reason for this lying in the hitherto lack of an adequate defect-chemical and nanoionic picture.

We report a synthetic route to achieve high electron mobility at room temperature in epitaxial La:BaSnO$_3$/SrZrO$_3$ heterostructures prepared on several oxide substrates. Room-temperature mobilities of 157, 145, and 143 cm$^2$V$^{-1}$s$^{-1}$ are achieved for heterostructures grown on DyScO$_3$ (110), MgO (001), and TbScO$_3$ (110) crystalline substrates, respectively. This is realized by first employing pulsed laser deposition to grow at very high temperature the SrZrO$_3$ buffer layer to reduce dislocation density in the active layer, then followed by the epitaxial growth of an overlaying La:BaSnO$_3$ active layer by molecular-beam epitaxy. Structural properties of these heterostructures are investigated, and the extracted upper limit of threading dislocations is well below $1.0\times 10^{10}$cm$^{-2}$ for buffered films on DyScO$_3$, MgO, and TbScO$_3$ substrates. The present results provide a promising route towards achieving high mobility in buffered La:BaSnO$_3$ films prepared on most, if not all, oxide substrates with large compressive or tensile lattice mismatches to the film., Comment: 9 pages and 5 Figures in the Maintext. 3 pages and 3 Figures in the Supplementary Information

The use of deep neural models for diagnosis prediction from clinical text has shown promising results. However, in clinical practice such models must not only be accurate, but provide doctors with interpretable and helpful results. We introduce ProtoPatient, a novel method based on prototypical networks and label-wise attention with both of these abilities. ProtoPatient makes predictions based on parts of the text that are similar to prototypical patients - providing justifications that doctors understand. We evaluate the model on two publicly available clinical datasets and show that it outperforms existing baselines. Quantitative and qualitative evaluations with medical doctors further demonstrate that the model provides valuable explanations for clinical decision support., Comment: AACL-IJCNLP 2022 Main Conference (Long Paper)

Recent work in natural language processing (NLP) has yielded appealing results from scaling model parameters and training data; however, using only scale to improve performance means that resource consumption also grows. Such resources include data, time, storage, or energy, all of which are naturally limited and unevenly distributed. This motivates research into efficient methods that require fewer resources to achieve similar results. This survey synthesizes and relates current methods and findings in efficient NLP. We aim to provide both guidance for conducting NLP under limited resources, and point towards promising research directions for developing more efficient methods., Comment: Accepted at TACL, pre publication version

Clinical phenotyping enables the automatic extraction of clinical conditions from patient records, which can be beneficial to doctors and clinics worldwide. However, current state-of-the-art models are mostly applicable to clinical notes written in English. We therefore investigate cross-lingual knowledge transfer strategies to execute this task for clinics that do not use the English language and have a small amount of in-domain data available. We evaluate these strategies for a Greek and a Spanish clinic leveraging clinical notes from different clinical domains such as cardiology, oncology and the ICU. Our results reveal two strategies that outperform the state-of-the-art: Translation-based methods in combination with domain-specific encoders and cross-lingual encoders plus adapters. We find that these strategies perform especially well for classifying rare phenotypes and we advise on which method to prefer in which situation. Our results show that using multilingual data overall improves clinical phenotyping models and can compensate for data sparseness., Comment: LREC 2022 submmision: January 2022

The mixed-valent spinel LiV$_{2}$O$_{4}$ is known as the first oxide heavy-fermion system. There is a general consensus that a subtle interplay of charge, spin, and orbital degrees of freedom of correlated electrons plays a crucial role in the enhancement of quasi-particle mass, but the specific mechanism has remained yet elusive. A charge-ordering (CO) instability of V$^{3+}$ and V$^{4+}$ ions that is geometrically frustrated by the V pyrochlore sublattice from forming a long-range CO down to $T$ = 0 K has been proposed as a prime candidate for the mechanism. To uncover the hidden CO instability, we applied epitaxial strain from a substrate on single-crystalline thin films of LiV$_{2}$O$_{4}$. Here we show a strain-induced crystallization of heavy fermions in a LiV$_{2}$O$_{4}$ film on MgO, where a charge-ordered insulator comprising of a stack of V$^{3+}$ and V$^{4+}$ layers along [001], the historical Verwey-type ordering, is stabilized by the in-plane tensile and out-of-plane compressive strains from the substrate. Our discovery of the [001] Verwey-type CO, together with previous realizations of a distinct [111] CO, evidence the close proximity of the heavy-fermion state to degenerate CO states mirroring the geometrical frustration of the pyrochlore lattice, which supports the CO instability scenario for the mechanism behind the heavy-fermion formation.

For La-doped BaSnO$_\text{3}$ thin films grown by pulsed laser deposition, we combine chemical surface characterization and electronic transport studies to probe the evolution of electronic states in the band structure for different La-doping content. Systematic analyses of spectroscopic data based on fitting the core electron line shapes help to unravel the composition of the surface as well as the dynamics associated with increasing doping. This dynamics is observed with a more pronounced signature in the Sn 3d core level, which exhibits an increasing asymmetry to the high binding energy side of the peak with increasing electron density. The present results expand the current understanding of the interplay between the doping concentration, electronic band structure and transport properties of epitaxial La:BaSnO$_\text{3}$ films., Comment: 7 Figures, 4 Tables in manuscript; and 6 Figures and 1 Table in the Supplementary Information

Superconductivity can be destroyed by a magnetic field with an upper bound known as the Pauli-limit in spin-singlet superconductors. Almost all the discovered superconductors are spin-singlet, with the highest transition temperature $T_c$ at ambient pressure achieved in the cuprate family. The closest cuprate analogue is the recently discovered infinite-layer nickelate, which hosts substantial structural and electronic similarity to the cuprate. A previous magnetotransport study on Nd$_{0.775}$Sr$_{0.225}$NiO$_2$ has observed an isotropic Pauli-limited upper critical field. Here, we report a large violation (>2 times) of Pauli-limit in every crystallographic directions in La$_{1-x}$(Ca/Sr)$_x$NiO$_2$ regardless of the doping $x$. Such a large violation of the Pauli-limit in all directions in La$_{1-x}$(Ca/Sr)$_x$NiO$_2$ is unexpected and unlikely accounted by a Fulde Ferrell-Larkin-Ovchinnikov (FFLO)-state, strong spin-orbit-coupling, strong-coupling or a large pseudogap. On the other hand, in agreement with the previous report, we observe a Pauli-limiting critical field in Nd$_{1-x}$Sr$_x$NiO$_2$ and the superconducting anisotropy decreases as doping increases, suggesting a spin-singlet pairing. Therefore, superconductivity in La$_{1-x}$(Ca/Sr)$_x$NiO$_2$ could be driven by a non-spin-singlet Cooper pairing mechanism with an attractive high-$T_c$ at 10 K, an order of magnitude higher than the known spin triplet superconductors, favourably extending the application of spin-triplet superconductivity in topological matter, non-dissipative spintronics, and quantum computing., Comment: 21 pages, 4 main figures, 2 extended data

Electron transparent TEM lamella with unaltered microstructure and chemistry is the prerequisite for successful TEM explorations. Currently, TEM specimen preparation of quantum nanostructures, such as quantum dots (QDs), remains a challenge. In this work, we optimize the sample-preparation routine for achieving high-quality TEM specimens consisting of SrRuO3 (SRO) QDs grown on SrTiO3 (STO) substrates. We demonstrate that a combination of ion-beam-milling techniques can produce higher-quality specimens of quantum nanostructures compared to TEM specimens prepared by a combination of tripod polishing followed by Ar+ ion milling. In the proposed method, simultaneous imaging in a focused ion-beam device enables accurate positioning of the QD regions and assures the presence of dots in the thin lamella by cutting the sample inclined by 5{\deg} relative to the dots array. Furthermore, the preparation of TEM lamellae with several large electron-transparent regions that are separated by thicker walls effectively reduces the bending of the specimen and offers broad thin areas. The final use of a NanoMill efficiently removes the amorphous layer without introducing any additional damage., Comment: 20 pages, 6 figures, 1 table

Artificial perovskite-oxide nanostructures possess intriguing magnetic properties due to their tailorable electron-electron interactions, which are extremely sensitive to the oxygen coordination environment. To date, perovskite-oxide nanodots with sizes below 50 nm have rarely been reported. Furthermore, the oxygen octahedral distortion and its relation to magnetic properties in perovskite oxide nanodots remain unexplored yet. Here, we have studied the magnetic anisotropy in patterned SrRuO3 (SRO) nanodots as small as 30 nm while performing atomic-resolution electron microscopy and spectroscopy to directly visualize the constituent elements, in particular oxygen ions. We observe that the magnetic anisotropy and RuO6 octahedra distortion in SRO nanodots are both nanodots' size-dependent but remain unchanged in the first 3-unit-cell interfacial SRO monolayers regardless of the dots' size. Combined with the first principle calculations, we unravel a unique structural mechanism behind the nanodots' size-dependent magnetic anisotropy in SRO nanodots, sugguesting that the competition between lattice anisotropy and oxygen octahedral rotation mediates anisotropic exchange interactions in SRO nanodots. These findings demonstrate a new avenue towards tuning magnetic properties of correlated perovskite oxides and imply that patterned nanodots could be a promising playground for engineering emergent functional behaviors., Comment: 21 pages and 4 figures for the main manuscript

Highly accurate alpha blending can be performed entirely with integer operations, and no divisions. To reduce the number of integer multiplications, multiple color components can be blended in parallel in the same 32-bit or 64-bit register. This tutorial explains how to avoid division operations when alpha blending with 32-bit RGBA pixels. An RGBA pixel contains four 8-bit components (red, green, blue, and alpha) whose values range from 0 to 255. Alpha blending requires multiplication of the color components by an alpha value, after which (for greatest accuracy) each of these products is divided by 255 and then rounded to the nearest integer. This tutorial presents an approximate alpha-blending formula that replaces the division operation with an integer shift and add -- and also enables the number of multiplications to be reduced. When the same blending calculation is carried out to high precision using double-precision floating-point division operations, the results are found to exactly match those produced by this approximation. C++ code examples are included., Comment: 10 pages, 1 figure

The superconducting infinite-layer nickelate family has risen as a promising platform for revealing the mechanism of high-temperature superconductivity. However, its challenging material synthesis has obscured effort in understanding the nature of its ground state and low-lying excitations, which is a prerequisite for identifying the origin of the Cooper pairing in high-temperature superconductors. In particular, the superconducting gap symmetry of nickelates has hardly been investigated and remains controversial. Here, we report the pairing symmetry of the infinite-layer nickelates determined by London penetration depth measurements in neodymium-based (Nd,Sr)NiO$_2$ and lanthanide-based (La,Ca)NiO$_2$ thin films of high crystallinity. A rare-earth-specific order parameter is observed. While the lanthanide nickelates follow dirty line-node behaviour, the neodymium-counterpart exhibits nodeless order parameters such as the $(d+is)$ wave. In contrast to the cuprates, our results suggest that the superconducting order parameter in nickelates is beyond a single $d_(x^2-y^2 )$-wave gap. Furthermore, the superfluid density shows a long tail near the superconducting transition temperature which is consistent with the emergence of a two-dimensional to three-dimensional crossover in the superconducting state. These observations challenge the early theoretical framework and propel further experimental and theoretical interests in the pairing nature of the infinite-layer nickelate family., Comment: Main manuscript: 26 pages, 3 figures; Supplementary file: 11 pages, 3 figures

Heterostructuring provides different ways to manipulate the orbital degrees of freedom and to tailor orbital occupations in transition metal oxides. However, the reliable prediction of these modifications remains a challenge. Here, we present a detailed investigation of the relationship between the crystal and electronic structure in YVO$_3$-LaAlO$_3$ superlattices by combining ab initio theory, scanning transmission electron microscopy, and x-ray diffraction. Density functional theory simulations including an on-site Coulomb repulsion term, accurately predict the crystal structure and in conjunction with x-ray diffraction, provide an explanation for the lifting of degeneracy of the vanadium $d_{xz}$ and $d_{yz}$ orbitals, that was recently observed in this system. In addition, we unravel the combined effects of electronic confinement and octahedral connectivity by disentangling their impact from that of epitaxial strain. Our results demonstrate that the specific orientation of the substrate and the thickness of the YVO$_3$ slabs in the multilayer, can be utilized to reliably engineer orbital polarization., Comment: 10 pages, 6 figures

Decision support systems based on clinical notes have the potential to improve patient care by pointing doctors towards overseen risks. Predicting a patient's outcome is an essential part of such systems, for which the use of deep neural networks has shown promising results. However, the patterns learned by these networks are mostly opaque and previous work revealed flaws regarding the reproduction of unintended biases. We thus introduce an extendable testing framework that evaluates the behavior of clinical outcome models regarding changes of the input. The framework helps to understand learned patterns and their influence on model decisions. In this work, we apply it to analyse the change in behavior with regard to the patient characteristics gender, age and ethnicity. Our evaluation of three current clinical NLP models demonstrates the concrete effects of these characteristics on the models' decisions. They show that model behavior varies drastically even when fine-tuned on the same data and that allegedly best-performing models have not always learned the most medically plausible patterns., Comment: NeurIPS 2021 Research2Clinics Workshop, Bridging the Gap: From Machine Learning Research to Clinical Practice

Topotactic transformations between related crystal structures are a powerful emerging route for the synthesis of novel quantum materials. Whereas most such "soft chemistry" experiments have been carried out on polycrystalline powders or thin films, the topotactic modification of single crystals -- the gold standard for physical property measurements on quantum materials -- has proven remarkably difficult. Here, we report the topotactic reduction of La$_{1-x}$Ca$_{x}$NiO$_3$ single-crystals to La$_{1-x}$Ca$_{x}$NiO$_{2+\delta}$ using CaH$_2$ as the reducing agent. The transformation from the three-dimensional perovskite to the quasi-two-dimensional infinite-layer phase was thoroughly characterized by X-ray diffraction, electron microscopy, Raman spectroscopy, magnetometry, and electrical transport measurements. Our work demonstrates that the infinite-layer structure can be realized as a bulk phase in crystals with micrometer-sized single-domains. The electronic properties of these specimens resemble those of epitaxial thin films rather than powders with similar compositions.

Heteroepitaxy offers a new type of control mechanism for the crystal structure, the electronic correlations, and thus the functional properties of transition-metal oxides. Here, we combine electrical transport measurements, high-resolution scanning transmission electron microscopy (STEM), and density functional theory (DFT) to investigate the evolution of the metal-to-insulator transition (MIT) in NdNiO$_3$ films as a function of film thickness and NdGaO$_3$ substrate crystallographic orientation. We find that for two different substrate facets, orthorhombic (101) and (011), modifications of the NiO$_6$ octahedral network are key for tuning the transition temperature $T_{\text{MIT}}$ over a wide temperature range. A comparison of films of identical thickness reveals that growth on [101]-oriented substrates generally results in a higher $T_{\text{MIT}}$, which can be attributed to an enhanced bond-disproportionation as revealed by the DFT+$U$ calculations, and a tendency of [011]-oriented films to formation of structural defects and stabilization of non-equilibrium phases. Our results provide insights into the structure-property relationship of a correlated electron system and its evolution at microscopic length scales and give new perspectives for the epitaxial control of macroscopic phases in metal-oxide heterostructures., Comment: 9 pages, 5 figures

Outcome prediction from clinical text can prevent doctors from overlooking possible risks and help hospitals to plan capacities. We simulate patients at admission time, when decision support can be especially valuable, and contribute a novel admission to discharge task with four common outcome prediction targets: Diagnoses at discharge, procedures performed, in-hospital mortality and length-of-stay prediction. The ideal system should infer outcomes based on symptoms, pre-conditions and risk factors of a patient. We evaluate the effectiveness of language models to handle this scenario and propose clinical outcome pre-training to integrate knowledge about patient outcomes from multiple public sources. We further present a simple method to incorporate ICD code hierarchy into the models. We show that our approach improves performance on the outcome tasks against several baselines. A detailed analysis reveals further strengths of the model, including transferability, but also weaknesses such as handling of vital values and inconsistencies in the underlying data., Comment: EACL 2021

Ion conducting materials are critical components of batteries, fuel cells, and devices such as memristive switches. Analytical tools are therefore sought that allow the behavior of ions in solids to be monitored and analyzed with high spatial resolution and in real time. In principle, inelastic tunneling spectroscopy offers these capabilities. However, as its spectral resolution is limited by thermal softening of the Fermi-Dirac distribution, tunneling spectroscopy is usually constrained to cryogenic temperatures. This constraint would seem to render tunneling spectroscopy useless for studying ions in motion. We report here the first inelastic tunneling spectroscopy studies above room temperature. For these measurements, we have developed high-temperature-stable tunnel junctions that incorporate within the tunnel barrier ultrathin layers for efficient proton conduction. By analyzing the vibrational modes of O-H bonds in BaZrO3-based heterostructures, we demonstrate the detection of protons with a spectral resolution of 20 meV at 400 K (FWHM). Overturning the hitherto existing prediction for the spectral resolution limit of 186 meV (5.4 kBT at 400 K), this resolution enables high-temperature tunneling spectroscopy of ion conductors. With these advances, inelastic tunneling spectroscopy constitutes a novel, valuable analytical tool for solid-state ionics., Comment: 5 figures in main text, 7 figures in supporting info

Explainability and interpretability are two important concepts, the absence of which can and should impede the application of well-performing neural networks to real-world problems. At the same time, they are difficult to incorporate into the large, black-box models that achieve state-of-the-art results in a multitude of NLP tasks. Bidirectional Encoder Representations from Transformers (BERT) is one such black-box model. It has become a staple architecture to solve many different NLP tasks and has inspired a number of related Transformer models. Understanding how these models draw conclusions is crucial for both their improvement and application. We contribute to this challenge by presenting VisBERT, a tool for visualizing the contextual token representations within BERT for the task of (multi-hop) Question Answering. Instead of analyzing attention weights, we focus on the hidden states resulting from each encoder block within the BERT model. This way we can observe how the semantic representations are transformed throughout the layers of the model. VisBERT enables users to get insights about the model's internal state and to explore its inference steps or potential shortcomings. The tool allows us to identify distinct phases in BERT's transformations that are similar to a traditional NLP pipeline and offer insights during failed predictions., Comment: Published in WWW '20: Companion Proceedings of the Web Conference 2020

Topological transport phenomena in magnetic materials are a major topic of current condensed matter research. One of the most widely studied phenomena is the ``topological Hall effect'' (THE), which is generated via spin-orbit interactions between conduction electrons and topological spin textures such as skyrmions. We report a comprehensive set of Hall effect and magnetization measurements on epitaxial films of the prototypical ferromagnetic metal SrRuO$_3$ the magnetic and transport properties of which were systematically modulated by varying the concentration of Ru vacancies. We observe Hall effect anomalies that closely resemble signatures of the THE, but a quantitative analysis demonstrates that they result from inhomogeneities in the ferromagnetic magnetization caused by a non-random distribution of Ru vacancies. As such inhomogeneities are difficult to avoid and are rarely characterized independently, our results call into question the identification of topological spin textures in numerous prior transport studies of quantum materials, heterostructures, and devices. Firm conclusions regarding the presence of such textures must meet stringent conditions such as probes that couple directly to the non-collinear magnetization on the atomic scale.

Hyperbolic materials exhibit unique properties that enable a variety of intriguing applications in nanophotonics. The topological insulator Bi2Se3 represents a natural hyperbolic optical medium, both in the THz and visible range. Here, using cathodoluminescence spectroscopy and electron energy-loss spectroscopy, we demonstrate that Bi2Se3, in addition to being a hyperbolic material, supports room-temperature exciton polaritons. Moreover, we explore the behavior of hyperbolic edge exciton polaritons in Bi2Se3. Edge polaritons are hybrid modes that result from the coupling of the polaritons bound to the upper and lower edges of Bi2Se3 nanoplatelets. In particular, we use electron energy-loss spectroscopy to compare Fabry-P\'erot-like resonances emerging in edge polariton propagation along pristine and artificially structured edges of the nanoplatelets. The experimentally observed scattering of edge polaritons by defect structures was found to be in good agreement with finite-difference time-domain simulations. Moreover, we experimentally proved coupling of localized polaritons in identical open and closed circular nanocavities to the propagating edge polaritons. Our findings are testimony to the extraordinary capability of the hyperbolic polariton propagation to cope with the presence of defects. This provides an excellent basis for applications such as nanooptical circuitry, cloaking at the nanometer scale, as well as nanoscopic quantum technology on the nanoscale.

This paper describes a 2-D graphics algorithm that uses shifts and adds to precisely plot a series of points on an ellipse of any shape and orientation. The algorithm can also plot an elliptic arc that starts and ends at arbitrary angles. The ellipse algorithm described here is largely based on earlier papers by Van Aken and Simar [1,2], which extend Marvin Minsky's well-known circle algorithm [3,4,5] to ellipses, and show how to cancel out the sources of error in Minsky's original algorithm. A new flatness test is presented for automatically controlling the spacing between points plotted on an ellipse or elliptic arc. Most of the calculations performed by the ellipse algorithm and flatness test use fixed-point addition and shift operations, and thus are well-suited to run on less-powerful processors. C++ source code listings are included. Keywords: parametric ellipse algorithm, rotated ellipse, Minsky circle algorithm, flatness, elliptic arc, conjugate diameters, affine invariance, Comment: 32 pages, 7 figures, C++ source code

We present Contextual Discourse Vectors (CDV), a distributed document representation for efficient answer retrieval from long healthcare documents. Our approach is based on structured query tuples of entities and aspects from free text and medical taxonomies. Our model leverages a dual encoder architecture with hierarchical LSTM layers and multi-task training to encode the position of clinical entities and aspects alongside the document discourse. We use our continuous representations to resolve queries with short latency using approximate nearest neighbor search on sentence level. We apply the CDV model for retrieving coherent answer passages from nine English public health resources from the Web, addressing both patients and medical professionals. Because there is no end-to-end training data available for all application scenarios, we train our model with self-supervised data from Wikipedia. We show that our generalized model significantly outperforms several state-of-the-art baselines for healthcare passage ranking and is able to adapt to heterogeneous domains without additional fine-tuning., Comment: The Web Conference 2020 (WWW '20)

The synthesis of stoichiometric Sr$_2$RuO$_4$ thin films has been a challenge because of the high volatility of ruthenium oxide precursors, which gives rise to ruthenium vacancies in the films. Ru vacancies greatly affect the transport properties and electronic phase behavior of Sr$_2$RuO$_4$, but their direct detection is difficult due to their atomic dimensions and low concentration. We applied polarized X-ray absorption spectroscopy at the oxygen K-edge and confocal Raman spectroscopy to Sr$_2$RuO$_4$ thin films synthesized under different conditions. The results show that these methods can serve as sensitive probes of the electronic and vibrational properties of Ru vacancies, respectively. The intensities of the vacancy-related spectroscopic features extracted from these measurements are well correlated with the transport properties of the films. The methodology introduced here can thus help to understand and control the stoichiometry and transport properties in films of Sr$_2$RuO$_4$ and other ruthenates.

The exchange bias effect is an essential component of magnetic memory and spintronic devices. Whereas recent research has shown that anisotropies perpendicular to the device plane provide superior stability against thermal noise, it has proven remarkably difficult to realize perpendicular exchange bias in thin-film structures. Here we demonstrate a strong perpendicular exchange bias effect in heterostructures of the quasi-two-dimensional canted antiferromagnet La$_2$CuO$_4$ and ferromagnetic (La,Sr)MnO$_3$ synthesized by ozone-assisted molecular beam epitaxy. The magnitude of this effect can be controlled via the doping level of the cuprate layers. Canted antiferromagnetism of layered oxides is thus a new and potentially powerful source of uniaxial anisotropy in magnetic devices., Comment: 7 pages, 5 figures

Bidirectional Encoder Representations from Transformers (BERT) reach state-of-the-art results in a variety of Natural Language Processing tasks. However, understanding of their internal functioning is still insufficient and unsatisfactory. In order to better understand BERT and other Transformer-based models, we present a layer-wise analysis of BERT's hidden states. Unlike previous research, which mainly focuses on explaining Transformer models by their attention weights, we argue that hidden states contain equally valuable information. Specifically, our analysis focuses on models fine-tuned on the task of Question Answering (QA) as an example of a complex downstream task. We inspect how QA models transform token vectors in order to find the correct answer. To this end, we apply a set of general and QA-specific probing tasks that reveal the information stored in each representation layer. Our qualitative analysis of hidden state visualizations provides additional insights into BERT's reasoning process. Our results show that the transformations within BERT go through phases that are related to traditional pipeline tasks. The system can therefore implicitly incorporate task-specific information into its token representations. Furthermore, our analysis reveals that fine-tuning has little impact on the models' semantic abilities and that prediction errors can be recognized in the vector representations of even early layers., Comment: Accepted at CIKM 2019

Magnetic ordering phenomena have a profound influence on the macroscopic properties of correlated-electron materials, but their realistic prediction remains a formidable challenge. An archetypical example is the ternary nickel oxide system RNiO3 (R = rare earth), where the period-four magnetic order with proposals of collinear and non-collinear structures and the amplitude of magnetic moments on different Ni sublattices have been subjects of debate for decades. Here we introduce an elementary model system - NdNiO3 slabs embedded in a non-magnetic NdGaO3 matrix - and use polarized resonant x-ray scattering (RXS) to show that both collinear and non-collinear magnetic structures can be realized, depending on the slab thickness. The crossover between both spin structures is correctly predicted by density functional theory and can be qualitatively understood in a low-energy spin model. We further demonstrate that the amplitude ratio of magnetic moments in neighboring NiO6 octahedra can be accurately determined by RXS in combination with a correlated double cluster model. Targeted synthesis of model systems with controlled thickness and synergistic application of polarized RXS and ab-initio theory thus provide new perspectives for research on complex magnetism, in analogy to two-dimensional materials created by exfoliation., Comment: This is the accepted version of the manuscript. The revised manuscript according to peer reviews was published online in Nature Physics on July 23, 2018