Search

As the feature size of integrated circuits continues to decrease, optical proximity correction (OPC) has emerged as a crucial resolution enhancement technology for ensuring high printability in the lithography process. Recently, level set-based inverse lithography technology (ILT) has drawn considerable attention as a promising OPC solution, showcasing its powerful pattern fidelity, especially in advanced process. However, massive computational time consumption of ILT limits its applicability to mainly correcting partial layers and hotspot regions. Deep learning (DL) methods have shown great potential in accelerating ILT. However, lack of domain knowledge of inverse lithography limits the ability of DL-based algorithms in process window (PW) enhancement and etc. In this paper, we propose an inverse lithography physics-informed deep neural level set (ILDLS) approach for mask optimization. This approach utilizes level set based-ILT as a layer within the DL framework and iteratively conducts mask prediction and correction to significantly enhance printability and PW in comparison with results from pure DL and ILT. With this approach, computation time is reduced by a few orders of magnitude versus ILT. By gearing up DL with knowledge of inverse lithography physics, ILDLS provides a new and efficient mask optimization solution.

${\gamma}$-GeSe is a new type of layered bulk material that was recently successfully synthesized. By means of density functional theory first-principles calculations, we systematically studied the physical properties of two-dimensional (2D) few-layer ${\gamma}$-GeSe. It is found that few-layer ${\gamma}$-GeSe are semiconductors with band gaps decreasing with increasing layer number; and 2D ${\gamma}$-GeSe with layer number $n \geq 2$ are ferroelectric with rather low transition barriers, consistent with the sliding ferroelectric mechanism. Particularly, spin-orbit coupling induced spin splitting is observed at the top of the valence band, which can be switched by the ferroelectric reversal; furthermore, their negative piezoelectricity also enables the regulation of spin splitting by strain. Finally, excellent optical absorption was also revealed. These intriguing properties make 2D few-layer ${\gamma}$-GeSe promising in spintronic and optoelectric applications., Comment: 18 pages,6 figures

Task-oriented dialog(TOD) aims to assist users in achieving specific goals through multi-turn conversation. Recently, good results have been obtained based on large pre-trained models. However, the labeled-data scarcity hinders the efficient development of TOD systems at scale. In this work, we constructed a weakly supervised dataset based on a teacher/student paradigm that leverages a large collection of unlabelled dialogues. Furthermore, we built a modular dialogue system and integrated coarse-to-fine grained classification for user intent detection. Experiments show that our method can reach the dialog goal with a higher success rate and generate more coherent responses., Comment: Towards Semi-Supervised and Reinforced Task-Oriented Dialog Systems Co-located with EMNLP 2022, System Description Paper, 5 pages

A group of newly discovered non-magnetic metal kagome structures AV3Sb5 (A=K, Rb, Cs) have aroused widespread interest in experiment and theory due to their unusual charge density wave (CDW) and intertwined superconductivity. However, they all possess weak electron-phonon coupling (EPC) and low superconducting transition temperature. Here, we performed high-throughput first-principles calculations on novel kagome candidates with AV3Sb5 prototype structure, and proposed 24 dynamically stable novel kagome metals. The calculation based on Bardeen-Cooper-Schrieffer theory shows that most of these metals are superconductors with much stronger EPC than the reported AV3Sb5 materials, and their superconducting transition temperatures Tc is between 0.3 and 5.0K. Additionally, several compounds, such as KZr3Pb5 with the highest Tc, are identified as Z2 topological metals with clear Dirac cone topological surface states near Fermi level. NaZr3As5 is shown to have possible CDW phases. Our results provide rich platforms for exploring various new physics with the prototype kagome structure, in which the coexistence of superconductivity and nontrivial topological nature provides promising insights on the discovery of topological superconductors.

Two-dimensional (2D) ferromagnetic and ferroelectric materials attract unprecedented attention due to the spontaneous-symmetry-breaking induced novel properties and multifarious potential applications. Here we systematically investigate a large family (148) of 2D MGeX3 (M = metal elements, X = O/S/Se/Te) by means of the high-throughput first-principles calculations, and focus on their possible ferroic properties including ferromagnetism, ferroelectricity, and ferroelasticity. We discover eight stable 2D ferromagnets including five semiconductors and three half-metals, 21 2D antiferromagnets, and 11 stable 2D ferroelectric semiconductors including two multiferroic materials. Particularly, MnGeSe3 and MnGeTe3 are predicted to be room-temperature 2D ferromagnetic half metals with Tc of 490 and 308 K, respectively. It is probably for the first time that ferroelectricity is uncovered in 2D MGeX3 family, which derives from the spontaneous symmetry breaking induced by unexpected displacements of Ge-Ge atomic pairs, and we also reveal that the electric polarizations are in proportion to the ratio of electronegativity of X and M atoms, and IVB group metal elements are highly favored for 2D ferroelectricity. Magnetic tunnel junction and water-splitting photocatalyst based on 2D ferroic MGeX3 are proposed as examples of wide potential applications. The atlas of ferroicity in 2D MGeX3 materials will spur great interest in experimental studies and would lead to diverse applications., Comment: 25 pages, 6 figures

Two-dimensional (2D) Janus transitional metal dichalcogenides (TMDCs) have great potential for photocatalytic water splitting due to their novel properties induced by the unique out-of-plane asymmetric structures. Here, we systematically investigate the geometric, electronic and optical properties of 2D Janus MoSSe with titanium doping and vacancies to explore their synergistic effects on photocatalytic activity. We find that there are effective attractions between the substituted or adsorbed Ti atoms and S/Se vacancies. The Ti adatoms dramatically extend the light absorption range to infrared region. The S/Se vacancies coexisting with Ti adatoms will modulate the transition of photo-excited electrons, thereby enhancing the sunlight absorption. The Ti adatoms either existing alone or coexisting with vacancies introduce smaller lattice distortion compared to substituted Ti atoms and these Ti adatoms induce smaller effective mass of charge carriers. The configuration of S vacancy coexisting with Ti adatoms on Se-surface exhibits the most significant synergistic effects and best overall photocatalytic performance. Our work reveals the mechanism and effects induced by doping and vacancies coexisting in 2D Janus TMDCs, also propose a new practical strategy to improve the performance of 2D photocatalysts.

Beyond the conventional trial-and-error method, machine learning offers a great opportunity to accelerate the discovery of functional materials, but still often suffers from difficulties such as limited materials data and unbalanced distribution of target property. Here, we propose the ab initio Bayesian active learning method that combines active learning and high-throughput ab initio calculations to accelerate prediction of desired functional materials with the ultrahigh efficiency and accuracy. We apply it as an instance to a large family (3,119) of two-dimensional hexagonal binary compounds with unbalanced materials property, and accurately screen out the materials with maximal electric polarization and proper photovoltaic band gaps, respectively, whereas the computational costs are significantly reduced by only calculating a few tenths of possible candidates in comparison to the random search. This approach shows enormous advantages for the cases with unbalanced distributions of target property. It can be readily applied to seek for a broad range of advanced materials.