Search

Moir\'e superlattices are emerging as a new route for engineering strongly correlated electronic states in two-dimensional van der Waals heterostructures, as recently demonstrated in the correlated insulating and superconducting states in magic-angle twisted bilayer graphene and ABC trilayer graphene/boron nitride moir\'e superlattices. Transition metal dichalcogenide (TMDC) moir\'e heterostructures provide another exciting model system to explore correlated quantum phenomena, with the addition of strong light-matter interactions and large spin-orbital coupling. Here we report the optical detection of strongly correlated phases in semiconducting WSe2/WS2 moir\'e superlattices. Our sensitive optical detection technique reveals a Mott insulator state at one hole per superlattice site ({\nu} = 1), and surprising insulating phases at fractional filling factors {\nu} = 1/3 and 2/3, which we assign to generalized Wigner crystallization on an underlying lattice. Furthermore, the unique spin-valley optical selection rules of TMDC heterostructures allow us to optically create and investigate low-energy spin excited states in the Mott insulator. We reveal an especially slow spin relaxation lifetime of many microseconds in the Mott insulating state, orders-of-magnitude longer than that of charge excitations. Our studies highlight novel correlated physics that can emerge in moir\'e superlattices beyond graphene., Comment: 25 pages, 7 figures

The task of natural language inference (NLI) is to identify the relation between the given premise and hypothesis. While recent NLI models achieve very high performance on individual datasets, they fail to generalize across similar datasets. This indicates that they are solving NLI datasets instead of the task itself. In order to improve generalization, we propose to extend the input representations with an abstract view of the relation between the hypothesis and the premise, i.e., how well the individual words, or word n-grams, of the hypothesis are covered by the premise. Our experiments show that the use of this information considerably improves generalization across different NLI datasets without requiring any external knowledge or additional data. Finally, we show that using the coverage information is not only beneficial for improving the performance across different datasets of the same task. The resulting generalization improves the performance across datasets that belong to similar but not the same tasks.

This paper describes DBPal, a new system to translate natural language utterances into SQL statements using a neural machine translation model. While other recent approaches use neural machine translation to implement a Natural Language Interface to Databases (NLIDB), existing techniques rely on supervised learning with manually curated training data, which results in substantial overhead for supporting each new database schema. In order to avoid this issue, DBPal implements a novel training pipeline based on weak supervision that synthesizes all training data from a given database schema. In our evaluation, we show that DBPal can outperform existing rule-based NLIDBs while achieving comparable performance to other NLIDBs that leverage deep neural network models without relying on manually curated training data for every new database schema., Comment: arXiv admin note: text overlap with arXiv:1804.00401

Moir\'e superlattices provide a powerful way to engineer properties of electrons and excitons in two-dimensional van der Waals heterostructures. The moir\'e effect can be especially strong for interlayer excitons, where electrons and holes reside in different layers and can be addressed separately. In particular, it was recently proposed that the moir\'e superlattice potential not only localizes interlayer exciton states at different superlattice positions, but also hosts an emerging moir\'e quasi-angular momentum (QAM) that periodically switches the optical selection rules for interlayer excitons at different moir\'e sites. Here we report the observation of multiple interlayer exciton states coexisting in a WSe2/WS2 moir\'e superlattice and unambiguously determine their spin, valley, and moir\'e QAM through novel resonant optical pump-probe spectroscopy and photoluminescence excitation spectroscopy. We demonstrate that interlayer excitons localized at different moir\'e sites can exhibit opposite optical selection rules due to the spatially-varying moir\'e QAM. Our observation reveals new opportunities to engineer interlayer exciton states and valley physics with moir\'e superlattices for optoelectronic and valleytronic applications.

As of any other satellite images, LAPAN-A3/IPB multispectral images suffered from both geometric and radiometric distortions which need to be corrected. LAPAN as satellite owner has developed image preprocessing algorithm to process raw image into systematically corrected image. This research aims to evaluate the performance of the developed algorithm, particularly the performance of lens vignetting and band co-registration correction as well as the performance of direct image georeferencing. Lens vignetting distortion occurs on image was corrected by using pre-flight calibration data, while calculation of direct georeferencing was done by using satellite metadata of satellite position and attitude. Meanwhile, band co-registration correction was conducted based entirely on the image being processed using image matching approach. Based on several results and analysis which have been done, lens vignetting effects on image can be suppressed significantly from about 40 percent down to 10 percent, band coregistration error can be reduced to below 2-3 pixels in average, and the calculated direct georeferencing has 3000 meter accuracy. The results show that the developed image preprocessing algorithm has moderately good performance to process LAPAN-A3/IPB multispectral images., Comment: 10 pages, 16 figures, journal

Moir\'e superlattices provide a powerful tool to engineer novel quantum phenomena in two-dimensional (2D) heterostructures, where the interactions between the atomically thin layers qualitatively change the electronic band structure of the superlattice. For example, mini-Dirac points, tunable Mott insulator states, and the Hofstadter butterfly can emerge in different types of graphene/boron nitride moir\'e superlattices, while correlated insulating states and superconductivity have been reported in twisted bilayer graphene moir\'e superlattices. In addition to their dramatic effects on the single particle states, moir\'e superlattices were recently predicted to host novel excited states, such as moir\'e exciton bands. Here we report the first observation of moir\'e superlattice exciton states in nearly aligned WSe2/WS2 heterostructures. These moir\'e exciton states manifest as multiple emergent peaks around the original WSe2 A exciton resonance in the absorption spectra, and they exhibit gate dependences that are distinctly different from that of the A exciton in WSe2 monolayers and in large-twist-angle WSe2/WS2 heterostructures. The observed phenomena can be described by a theoretical model where the periodic moir\'e potential is much stronger than the exciton kinetic energy and creates multiple flat exciton minibands. The moir\'e exciton bands provide an attractive platform to explore and control novel excited state of matter, such as topological excitons and a correlated exciton Hubbard model, in transition metal dichalcogenides.

The Berry phase of Bloch states can have profound effects on electron dynamics lead to novel transport phenomena, such as the anomalous Hall effect and the valley Hall effect. Recently, it was predicted that the Berry phase effect can also modify the exciton states in transition metal dichalcogenide monolayers, and lift the energy degeneracy of exciton states with opposite angular momentum through an effective valley-orbital coupling. Here, we report the first observation and control of the Berry-phase induced splitting of the 2p-exciton states in monolayer molybdenum diselenide using the intraexciton optical Stark spectroscopy. We observe the time-reversal-symmetric analog of the orbital Zeeman effect resulting from the valley-dependent Berry phase, which leads to energy difference of +14 (-14) meV between the $2p^+$ and $2p^-$ exciton states in +K (-K) valley, consistent with the ordering from our ab initio GW-BSE results. In addition, we show that the light-matter coupling between intraexciton states are remarkably strong, leading to prominent valley-dependent Autler-Townes doublet under resonant driving. Our study opens up new pathways to coherently manipulate the quantum states and excitonic excitation with infrared radiation in two-dimensional semiconductors.