Abstract: A main and auxiliary alloy-based neodymium-iron-boron magnet material and the preparation method thereof. The raw material composition for the main and auxiliary alloy-based neodymium-iron-boron magnet material includes a main alloy raw material and an auxiliary alloy raw material, wherein the mass percentage of the auxiliary alloy raw material in the raw material composition for the main and auxiliary alloy-based neodymium-iron-boron magnet material is 1.0-15.0 mass %. For the main and auxiliary alloy-based neodymium-iron-boron magnet material prepared by using the raw material composition, the coercivity is increased while high remanence is ensured, and the preparation method therefor can be suitable for engineering applications.

Abstract: Provided is a preparation method of a detector material. The present disclosure epitaxially grows a buffer layer on a surface of a gallium arsenide substrate, deposits a silicon dioxide layer on the buffer layer, and etches the silicon dioxide layer on the buffer layer according to a strip pattern by photolithography and etching to form strip growth regions with continuous changes in width. Finally, a molecular beam epitaxy (MBE) technology is used to epitaxially grow the detector material in the strip growth regions under set epitaxy growth conditions. Because of the same mobility of atoms arriving at the surface of the substrate, numbers of atoms migrating to the strip growth regions are different due to different widths of the strip growth regions, such that compositions of the material change with the widths of the strip growth regions or a layer thickness changes with the widths of the strip growth regions.

Abstract: A method includes obtaining raw image data, where the raw image data includes data values each having most significant bits and least significant bits. The method also includes providing the raw image data to a trained machine learning model and generating processed image data using the trained machine learning model. The method further includes presenting an image based on the processed image data. The trained machine learning model is trained to modulate a feature map associated with the most significant bits of the data values of the raw image data based on the least significant bits of the data values of the raw image data in order to generate a fusion of the most significant bits and the least significant bits of the data values of the raw image data.

Abstract: Provided is a preparation method of a detector material. The present disclosure epitaxially grows a buffer layer on a surface of a gallium arsenide substrate, deposits a silicon dioxide layer on the buffer layer, and etches the silicon dioxide layer on the buffer layer according to a strip pattern by photolithography and etching to form strip growth regions with continuous changes in width. Finally, a molecular beam epitaxy (MBE) technology is used to epitaxially grow the detector material in the strip growth regions under set epitaxy growth conditions. Because of the same mobility of atoms arriving at the surface of the substrate, numbers of atoms migrating to the strip growth regions are different due to different widths of the strip growth regions, such that compositions of the material change with the widths of the strip growth regions or a layer thickness changes with the widths of the strip growth regions.

Abstract: A data processing method includes, in response to an interaction message being displayed on an interaction interface, displaying a virtual object in an interaction area in which the interaction message is located, and displaying a resource object at a target location on the interaction interface. The virtual object and the resource object are associated with the interaction message, and the resource object and the virtual object are configured to interact with each other. The method further includes displaying, on the interaction interface, interactive movement involving the virtual object and the resource object. The interactive movement includes moving the virtual object from a location of the interaction area, and/or moving the resource object from the target location.

Abstract: A method for generating adversarial attacks on a video recognition model is disclosed, including (a) generating the content of bullet-screen comments (BSCs) by an image captioning model for a clean video sample; (b) generating an adversarial video sample by inserting the BSCs into the clean video sample; (c) using the adversarial video sample to attack the video recognition model; (d) receiving rewards from the environment; (e) optimizing the position and transparency of BSCs by an reinforced learning (RL) agent according to the received rewards; (f) updating the adversarial video sample and using it to attack the video recognition model; and iteratively repeating steps of (d)-(f), until a predefined condition is matched.