Adaptive feature interaction enhancement network for text classification.

Text classification aims to establish text distinctions, which face difficulty in capturing global text semantics and local details. To address this issue, we propose an Adaptive Feature Interactive Enhancement Network (AFIENet). Specifically, AFIENet uses two branches to model the text globally and locally.

High-Quality Solution-Processed Quasi-2D Perovskite for Low-Threshold Lasers.

Spin-coated quasi-two-dimensional halide perovskite films, which exhibit superior optoelectronic properties and environmental stability, have recently been extensively studied for lasers. Crystallinity is of great importance for the laser performance. Although some parameters related to the spin-coating process have been studied, the in-depth understanding and effective control of the acceleration rate on two-dimensional perovskite crystallization during spin-coating are still unknown.

Localized Bound Multiexcitons in Engineered Quasi-2D Perovskites Grains at Room Temperature for Efficient Lasers.

Reducing the excitation threshold to minimize the Joule heating is critical for the realization of perovskite laser diodes. Although bound excitons are promising for low threshold laser, how to generate them at room temperature for laser applications is still unclear in quasi-2D perovskite-based devices. In this work, via engineering quasi-2D perovskite PEA (CH NH ) Pb Br microscopic grains by the anti-solvent method, room-temperature multiexciton radiative recombination is successfully demonstrated at a remarkably low pump density of 0.

Stabilization of the High-Energy-Density CuN Salts under Ambient Conditions by a Ligand Effect.

A series of excellent works have demonstrated that high-nitrogen-content metal pentazolate (N ) compounds could be stabilized by high pressure. However, under ambient conditions, low stability precludes their synthesis and application in the field of high-energy-density material. In this work, by using a constrained structure search method, we predicted two new structures as 222-CuN and 2/-CuN containing N with strong N-N and Cu-N bonds.