Abstract: A continuous recycling device of an organic base, including a first pump, a second pump, a third pump, a micro-mixer, a micro-channel reactor, a liquid-liquid separator and a continuous stirred tank. The third pump is configured to feed a first base-containing waste liquid to the micro-mixer. The first pump is configured to feed a solution of a second base to the micro-mixer. The micro-mixer is configured to mix the first base-containing waste liquid with the solution of the second base, and transfer the mixture to the micro-channel reactor. The liquid-liquid separator is configured to divide the reaction mixture into a top layer and a bottom layer. The continuous stirred tank is filled with a water-absorbing agent to absorb water in the top layer. A continuous recycling method using such device is also provided.

Abstract: A method for fully continuous-flow preparation of a nitroimidazole antimicrobial agent using a fully continuous-flow system is provided herein. The method adopts a raw material of a glyoxal aqueous solution, an acetaldehyde aqueous solution and an ammonium hydroxide solution followed by multi-step chemical reactions and continuous post-treatments to yield a nitroimidazole antimicrobial agent with a high purity. The fully continuous-flow system for implementing the method includes a plurality of micromixers, a plurality of microreactors, a plurality of back pressure valves, a plurality of gas-liquid separators, a first plurality of online solvent switching units and a feed unit.

Abstract: A fully-continuous synthesis method of cyproterone acetate is provided. 4-androstene-3,17-dione is adopted as a starting material. The present disclosure adopts a fully continuous device composed of micromixers, microreactors, online gravity separation units, online solvent switching units, online solvent concentration units and solvent recovery systems connected according to a cyproterone acetate synthesis route. Cyproterone acetate product is synthesized through one enzymatic catalytic reaction, nine chemical reactions and continuous operations. This can realize recovery of dichloromethane, dichloroethane and ethanol, and significantly reduce emission of three wastes. The obtained crude cyproterone acetate is subjected to decolorization, recrystallization, filtration and drying to obtain pure cyproterone acetate with a purity greater than 99%.

Abstract: A drug screening device is provided. A method of determining optimal drug concentrations and efficacy in a patient using the device are provided. A method of determining effective chemotherapeutic drugs and effective concentrations thereof using the device is provided. Also, a method of determining safety and efficacy of drugs using the device is provided.

Abstract: An efficient structure and methodology for container-based applications are provided for disaster recovery for data in a storage system using application consistency groups for replication from the storage system to a disaster recovery system. Application consistency groups can be generated containing applications, a disaster recovery plan, and a disaster recovery policy, where each application has a consistency level. The applications in an application consistency group can include stateless applications and stateful applications. The replication of an application consistency group from the system to a disaster recovery storage system can be controlled according to the disaster recovery policy, the disaster recovery plan, and the consistency levels of the application consistency group.

Abstract: A full continuous-flow preparation method of (+)-biotin, including: subjecting a cyclic anhydride and a chiral biphenyl propylene glycol to asymmetric ring-opening reaction to produce a first intermediate, which undergoes selective reduction with a borohydride and cyclization with an inorganic mineral acid to produce (3aS, 6aR)-lactone; subjecting the (3aS, 6aR)-lactone and a sulfenylating reagent to sulfenylation to produce (3aS, 6aR)-thiolactone, which undergoes Fukuyama coupling with a zinc reagent in the presence of a palladium catalyst and elimination reaction in the presence of an inorganic mineral acid to produce an alkenyl valerate compound; subjecting the alkenyl valerate compound to reduction in the presence of a Pd/C catalyst to produce a valerate ester, which undergoes hydrolysis to produce a valeric acid salt; and subjecting the valeric acid salt to debenzylation in the presence of an inorganic mineral acid to produce the target product (+)-biotin.

Abstract: A systematic security protection method for a multimodal AI model is provided. The method comprises: in a security detection stage of the target multimodal AI model, according to a perceptual distance between detection data and original data determined by a trained distance-aware judgment model, obtaining detection data whose perceptual distance from the original data meets a requirement, and detecting the target multimodal AI model by using the detection data; in a case that the detection result in the detection stage is that the target multimodal AI model has a security problem, on one hand, performing security coding protection and tokenization filtering on the input of the model, on the other hand, performing multi-dimensional omnidirectional evaluation on the multimodal AI model.

Abstract: Disclosed are recombinant vectors comprising one or more polynucleotides encoding at least one modified ribonucleic acid (RNA) dependent RNA polymerase (RdRp), as well as pharmaceutical compositions and methods of their use.

Abstract: A systematic security protection method for a multimodal AI model is provided. The method comprises: in a security detection stage of the target multimodal AI model, according to a perceptual distance between detection data and original data determined by a trained distance-aware judgment model, obtaining detection data whose perceptual distance from the original data meets a requirement, and detecting the target multimodal AI model by using the detection data; in a case that the detection result in the detection stage is that the target multimodal AI model has a security problem, on one hand, performing security coding protection and tokenization filtering on the input of the model, on the other hand, performing multi-dimensional omnidirectional evaluation on the multimodal AI model.

Abstract: Techniques disclosed herein enable training and/or utilizing speaker dependent (SD) speech models which are personalizable to any user of a client device. Various implementations include personalizing a SD speech model for a target user by processing, using the SD speech model, a speaker embedding corresponding to the target user along with an instance of audio data. The SD speech model can be personalized for an additional target user by processing, using the SD speech model, an additional speaker embedding, corresponding to the additional target user, along with another instance of audio data. Additional or alternative implementations include training the SD speech model based on a speaker independent speech model using teacher student learning.

Abstract: A fully-continuous synthesis method of glyphosate is provided, which is performed by using a fully-continuous system including a feed pump, a plurality of micromixers, a plurality of microchannel reactors, a dynamic rotary reactor, a buffer tank, a back pressure valve, a plurality of reaction vessels and a continuous crystallizer. Glycine is used as a raw material, and reacted with a paraformaldehyde depolymerization product to yield N, N-dihydroxymethylglycine, which further undergoes esterification with dimethyl phosphite to generate methyl glyphosate. The methyl glyphosate is adjusted to be acidic, desolvated, hydrolyzed and purified to obtain glyphosate.

Abstract: Disclosed are a paddle discarding fully-sealed cement stirring and solidification bucket and a liquid injection and stirring method. A stirring paddle, a bucket cover, and a stirring steel bucket are first assembled and conveyed to a feeding station for cover removal and cement filling; then the stirring steel bucket is conveyed to a stirring station; after a stirring power device, a waste liquid injection device, and a breathing and exhaust device are correspondingly connected to a stirring paddle drive connection end, a waste liquid injection interface, and a breathing and exhaust device interface, injection and stirring of radioactive waste liquid are carried out in a fully-sealed environment. In the present invention, the stirring paddle is discarded in the steel bucket, thereby avoiding the problems of secondary radioactive waste liquid produced by cleaning the stirring paddle and diffusion of radioactive aerosol in the stirring process.

Abstract: A device for fully-continuous synthesis of glyphosate is provided. The device includes a first feed pump, a second feed pump, a third feed pump, a fourth feed pump, a fifth feed pump, a six feed pump, a seventh feed pump, a first micromixer, a second micromixer, a third micromixer, a fourth micromixer, a first microchannel reactor, a second microchannel reactor, a third microchannel reactor, a fourth microchannel reactor, a fifth microchannel reactor, a sixth microchannel reactor, a first stirring vessel, a second stirring vessel, a third stirring vessel, a buffer tank, a first back pressure valve, a second back pressure valve and a continuous crystallizer that are connected according to a glyphosate synthesis route. The device can realize continuous production of glyphosate.

Abstract: A continuous-flow preparation method for an amino alcohol compound using a micro-reaction system. The micro-reaction system includes a feed pump, a micromixer, a microchannel reactor, and a back pressure valve. An aldehyde compound and an amine compound are simultaneously fed to the micro-mixer for mixing to obtain a mixed solution. The mixed solution is directly fed to the micro-channel reactor and undergoes an addition reaction. The reaction mixture is collected, and subjected to concentration, separation and purification to obtain the desired amino alcohol compound.

Abstract: A computer implemented method comprises accessing data stored in memory of the computer system, wherein the memory includes a faster access storage medium and a slower access storage medium; determining respective exponential moving averages of a number of access operations performed on respective memory locations of the memory; and transferring a portion of the data into the faster access storage medium and transferring a portion of the data into the slower access storage medium according to the determined exponential moving average.

Abstract: A method comprises forming a single-crystal-like metal layer on a metal seed layer, the metal seed layer formed on a carrier wafer. The method comprises forming a first diffusion layer on the single-crystal-like metal layer. The method also comprises forming a second diffusion layer on a dielectric layer of a target substrate, the target substrate comprising one or more recording head subassemblies. The method further comprises flipping and joining the carrier wafer with the target substrate such that the first and second diffusion layers are bonded and the single-crystal-like metal layer is integrated with the recording head as a near-field transducer.

Abstract: A method for continuous-flow amination for an alkyl carboxylic acid compound is provided, in which an amination reagent and a catalyst are mixed in a first micro-mixer and then preheated in a preheater. The mixture is mixed with a substituted alkyl carboxylic acid solution in a second micro-mixer and reacted in a dynamic flow reactor to produce a mixture including a carboxyl-containing organic amine product. Then, the mixture is subjected to gas-liquid separation, and the liquid phase is collected and filtered to obtain a first filtrate and a first filter residue. The first filter residue is mixed with a base solution under stirring and filtered to obtain a second filtrate and a second filter residue. The second filter residue was dried to yield a high-purity carboxyl-containing organic amine product.

Abstract: A method of preparing an alkyl phosphate compound based on a micro-reaction system. The micro-reaction system includes a feed pump, a first micro-mixer, a second micro-mixer, a first micro-channel reactor, a second micro-channel reactor and a back-pressure device, where the first micro-mixer, the second micro-mixer, the first micro-channel reactor, the second micro-channel reactor and the back-pressure device are sequentially connected. The method includes the following steps. An alkylamine compound and an acid-binding agent are simultaneously fed to a first micro-mixer for mixing and then to the first micro-channel reactor for pre-reaction to obtain a pre-reaction solution. The pre-reaction solution and a phosphate or phosphite are simultaneously fed to the second micro-mixer for mixing to obtain a first reaction mixture. The first reaction mixture is fed to the second micro-channel reactor to carry out a condensation reaction under a back pressure condition and is concentrated to obtain the final product.

Abstract: This disclosure provides synchronization methods and apparatuses for cross-device navigation tasks. One method includes: a first terminal obtains navigation data from a first navigation application through a first navigation transfer module in response to a preset cross-device transmission condition is satisfied. The first terminal generates navigation synchronization information of the navigation data through the first navigation transfer module. A second terminal receives the navigation synchronization information sent by the first terminal, and transmits the navigation synchronization information to a second navigation application through a second navigation transfer module, and updates a navigation task of a second navigation application based on the navigation synchronization information.

Abstract: Processing of acoustic features of audio data to generate one or more revised versions of the acoustic features, where each of the revised versions of the acoustic features isolates one or more utterances of a single respective human speaker. Various implementations generate the acoustic features by processing audio data using portion(s) of an automatic speech recognition system. Various implementations generate the revised acoustic features by processing the acoustic features using a mask generated by processing the acoustic features and a speaker embedding for the single human speaker using a trained voice filter model. Output generated over the trained voice filter model is processed using the automatic speech recognition system to generate a predicted text representation of the utterance(s) of the single human speaker without reconstructing the audio data.