Abstract: In a model training method, a computing device generates an initial virtual human speaking video based on an audio data set and a person speaking video. The computing device trains a lip synchronization parameter generation model based on the audio data set and a lip synchronization training parameter used as a model training label. The computing device then uses the lip synchronization parameter generation model to generate a lip synchronization parameter based on input audio, and the lip synchronization parameter is used to drive a virtual human to speak. The computing device expands a data amount of the lip synchronization training parameter by using the initial virtual human speaking video.

Abstract: A rendering model training method and apparatus are provided. A related video rendering method and apparatus, a device, and a storage medium are also provided. The rendering model training method includes: obtaining a first video including a face of a target object; mapping a facial action of the target object in the first video based on a three-dimensional face model, to obtain a second video including a three-dimensional face; and training an initial rendering model by using the second video as an input of the initial rendering model and using the first video as supervision of an output of the initial rendering model, to obtain a target rendering model. The second video generated based on the three-dimensional face model is used as a sample for training the rendering model.

Abstract: Machine-learning systems and methods can perform multi-label classification of medical images in a clinical context. A system can incorporate a triplet attention network that combines category-attention, self-attention, and cross-attention to learn high-quality label embeddings by mining effective information from medical images. The set of labels can include a single “hybrid” label assigned to multiple low-frequency diseases. The machine-learning system can be trained using a dual-pool contrastive learning technique. In the inference stage, the trained network can use a dual-pool contrastive inference technique to reduce the likelihood of false negatives and enhance the ability to detect diseases not seen in the training data.

Abstract: The invention relates to bone repair and more specifically to a berberine/mineralized collagen-based composite membrane, a preparation method, and an application thereof. The composite membrane includes a berberine nanofiber membrane and a mineralized collagen membrane disposed on a unilateral surface of the berberine nanofiber membrane. The mineralized collagen with biomimetic mineralization capacity is combined with a Chinese materia medica monomer, berberine, the resulting bilayer composite membrane has a better effect on promoting osteogenesis In addition, a novel dosage form of berberine is constructed. An electrospinning method prepares a berberine nanofiber membrane. The berberine nanofibers are received by the mineralized collagen membrane, or after the berberine nanofiber membrane is obtained, a mineralized collagen membrane is prepared by applying on the surface of the berberine nanofiber membrane.

Abstract: The present invention relates to a multi-component cross-container control method, comprising: displaying an editor page, the editor page comprising a canvas area, and a plurality of components in a first container being displayed in the canvas area; and in response to a dragging input for at least two components in the plurality of components, when at least one component of the at least two components is overlapped with a second container, adding at least two components of target components into the second container, the second container being a sub-container or a parent container of the first container.

Abstract: Disclosed is an antigen peptide with an infectious bronchitis virus (IBV)-specific neutralizing epitope, in which the peptide is a cyclic polypeptide, and the amino acid sequence of the peptide is CSCPYVSYGRFCIQPDGSIKQC. Also disclosed are an IBV specific antibody and a preparation method thereof. Further disclosed is an ELISA method as an alternative to the neutralization potency assay. The disclosure also discloses an ELISA detection kit, and using the established pELISA method to detect IBV antibodies, it is found that the method is positively correlated with anti-IBV neutralizing antibodies, which may be used to evaluate the immune effect of IBV vaccines on IBV infected chickens and to determine the antibody level, thereby benefiting the health management of chicken flocks.

Abstract: The present disclosure relates to the technical field of bone repair and provides a berberine/mineralized collagen-based composite membrane as well as a preparation method and an application thereof. The composite membrane of the present disclosure includes a berberine nanofiber membrane and a mineralized collagen membrane disposed on a unilateral surface of the berberine nanofiber membrane. In the present disclosure, the mineralized collagen with biomimetic mineralization capacity is combined with a Chinese materia medica monomer, berberine, the resulting bilayer composite membrane has a better effect on promoting osteogenesis; in addition, a novel dosage form of berberine is constructed.

Abstract: The present application provides a high-affinity PD-1 mutant polypeptide, and a fusion protein of the PD-1 mutant polypeptide and an antibody (specifically binding to a tumor antigen) and preparation and application thereof.

Abstract: The present disclosure relates to an antibody fragment specifically binding to PSMA, particularly to a single-chain Fv (scfv), a bispecific antibody containing the scFv, and a chimeric antigen receptor (CAR) and the preparation and the use thereof.

Abstract: The present disclosure provides an immunocytokine with IL-15 and IL-15Ra fused to an antibody targeting a tumor cell surface antigen, which can effectively and specifically target the complex of IL-15 and IL-15Ra to tumor microenvironment, and activate relevant immune cells within or in the vicinity of a tumor, thereby achieving the goal of tumor-specific killing, and at the same time, preventing immunotoxicity induced by the systemic hyperactivation of NK cells.

Abstract: A method for improving the uniformity of connection performance of an assembly surface based on zoned hardening is provided. Through a finite element contact analysis of the assembly surface, a topology optimization of the hardening layer layout of the assembly surface is carried out with the aim of improving the uniformity of the connection performance of the assembly surface. The design is based on the optimized theoretical data, and uses the laser hardening technology to realize the zoned differential hardening of the assembly surface. By performing the zoned differential hardening on the assembly surface, the uniformity of the connection performance of the assembly surface can be effectively improved, and the purpose of improving the high cycle fatigue and the vibration of the high-end equipment parts such as aero-engines can be further achieved.

Abstract: Provided are a bispecific molecule and preparation and use thereof. The bispecific molecule includes a molecule that specifically binds an interleukin-1 receptor (IL-1R) and an antibody that targets a free inflammatory factor. The molecule that specifically binds the cell surface interleukin-1 receptor (IL-1R) aggregates the antibody that targets the free inflammatory factor linked thereto on or near the cell surface, thereby the local concentration of the bispecific molecule on or near the cell surface is increased, adverse reactions are avoided, treatment effectiveness is increased and the infection risk of patients is also reduced.

Abstract: Provided are a novel bispecific binding molecule and use thereof. The novel bispecific binding molecule-drug conjugate includes: a) a bispecific binding molecule consisting of a first binding moiety binds a tumor cell surface antigen (T) and a second antigen binding moiety binds to an internalizing effector protein (E), and b) a cytotoxic ingredient covalently conjugated to the bispecific binding molecule. The bispecific binding molecule may specifically exert cytotoxic activity against tumor cells by specifically targeting to the tumor cells through the first antigen binding moiety, and internalizing the cytotoxic ingredient into the tumor cells by the second antigen binding moiety of the bispecific binding molecule.

Abstract: The disclosure provides a method for fast reconfiguration of power supply grid in tens of milliseconds after power grid failure. The master station of fast reconfiguration of power grid concentrates the network status information from the client stations at transformer substations or power plants, and compares it with the built-in control strategy table which deals with possible faults. When an expected power gird disconnection fault is detected, the pre-start switch-on instruction is sent to the client stations with multiple breakers which can reconnect the grid. When the fault is cleared, the client stations shall identify the fault clearing time according to the local information, and send the instruction of synchronous switch-on to the corresponding local breakers which can reconnect the separated grids.

Abstract: A method for preparing an intelligent antibacterial and antioxidative film involves preparing a PVA solution; adding nano-TiO2 to the PVA solution to obtain a PVA-TiO2 solution; determining the optimal amount of nano-TiO2; preparing a PSPC solution; preparing a PSPC-TiO2-PVA solution; and producing a PSPC/TiO2/PVA film. The film has better mechanical performance than saccharide and protein films. Shelf life of food is prolonged as the film possesses antibacterial and antioxidative properties. Furthermore, the film shows different colors in various pH environments. The film has a wide range of applications in food packaging owing to the integration of color development and antibacterial and antioxidative properties.