Abstract: Disclosed is a method for coffee bean defect detection based on a Siamese neural network. The method includes: obtaining image data of normal coffee beans and four types of defective coffee beans, performing a preprocessing operation on the image data, and dividing the preprocessed image data into a training set, a validation set, and a test set according to a preset ratio; constructing a coffee bean defect detection model based on a dual-branch Siamese neural network, and training the coffee bean defect detection model using the training set to obtain a similarity score; validating and adjusting the trained coffee bean defect detection model using the validation set by employing a mixed-precision training strategy and a dynamic learning rate adjustment technique; and performing defect detection inference on the validated and adjusted coffee bean defect detection model using the test set to complete the coffee bean defect detection.

Abstract: A liquid cooling system includes a thermoelectric cooler (TEC) between a radiator plate and a radiator, and a thermoelectric generator (TEG) at a location where the TEG is driven by heat from a chip package. The chip package is cooled by a cold plate of the liquid cooling system and the TEC is controlled by the TEG. The TEG may be between the chip package and the cold plate or elsewhere in or adjacent to the chip package. The TEG may control the TEC through a relay. The TEG automatically activates the TEC when the chip package is under peak load.

Abstract: Some aspects relate to technologies providing a framework for estimating content reach and reviewing personalized content. In accordance with some aspects, a user requests to estimate the reach of content variations of code-based content to target profiles. A background thread is initiated to display progressive results and representative content variations for the user to review. In order to determine the estimated reach, an abstract syntax tree is determined from the content and the target profiles are sampled. The number of profiles per content variation is determined based on the tree evaluation state of each of the sampled profiles. The estimated reach of each content variation is determined using an estimation technique based on the number of profiles per content variation and the total number of profiles. The estimated reach of each content variation is refined by increasing the sample size when sampling the target profiles.

Abstract: A secondary battery includes an electrode assembly and an electrolyte. The electrode assembly includes a positive electrode plate, a negative electrode plate, and a separator, where the positive electrode plate includes a positive electrode active material layer, the positive electrode active material layer including a positive electrode active material; and the negative electrode plate includes a negative electrode active material layer, the negative electrode active material layer including a negative electrode active material; a coating weight of the positive electrode active material layer is Wz, a coating weight of the negative electrode active material layer is Wf, and Wz and Wf satisfy: 1.6Wf?Wz?2.2Wf, and 3.25 mg/cm2?Wf?5.84 mg/cm2; and Dv99 of the positive electrode active material is 27 ?m to 33 ?m, and Dv99 of the negative electrode active material is 23 ?m to 28 ?m.

Abstract: One aspect of the present disclosure pertains to an integrated circuit (IC) structure. The IC structure includes a die and an integrated heat sink structure disposed over the die. In some embodiments, the integrated heat sink structure includes a first closed-loop microchannel structure adjacent to the die and a second closed-loop microchannel structure disposed over the first closed-loop microchannel structure. In an example, the second closed-loop microchannel structure is disposed further away from the die than the first closed-loop microchannel structure. In some implementations, a plurality of microchannels and a micromixer chamber collectively provide the first and second closed-loop microchannel structures.

Abstract: The present disclosure relates to a nucleic acid-lipid nanoparticle composition and a lyophilized preparation thereof, and further relates to a preparation method for the composition and a use of the composition. The nucleic acid-lipid nanoparticle composition of the present disclosure comprises lactate, so that the stability of nucleic acid-lipid nanoparticles is improved in a lyophilization process and a prepared lyophilized preparation.

Abstract: One aspect of the present disclosure pertains to an IC packaging structure that includes a bottom circuit structure having first semiconductor devices on a first substrate, a first interconnect structure over the first semiconductor devices, and a first bonding structure over the first interconnect structure; and a top circuit structure having second semiconductor devices on a second substrate, a second interconnect structure underlying the second semiconductor devices, and a second bonding structure underlying the second interconnect structure. The first bonding structure includes a first metal feature having a first crystalline bulk layer of a metal and a first amorphous surface layer of the metal. The second bonding structure includes a second metal feature having a second crystalline bulk layer of the metal and a second amorphous surface layer of the metal. The top circuit structure is bonded to the bottom circuit structure through the first and second amorphous surface layers.

Abstract: The present invention relates to a method for preparing a bisamide insecticide or an intermediate thereof. Specifically, the present invention relates to a method for preparing a compound of Formula I. The method comprises the step of reacting a carboxylic acid as represented by Formula I-1 with a halogenating agent and elemental halogen. The method can substantially improve product yield, producing relatively high economic benefits.

Abstract: Provided are a filter and a multiplexer. The filter includes a first port, a second port, and at least one filtering unit. Each of the at least one filtering unit includes a ?-type LC network and a resonator, the ?-type LC network includes at least one first element and at least two second elements. Each first element of the at least one first element is connected in series between the first port and the second port. A first end of at least one second element is connected between the first port and a first end of the first element. A first end of at least one second element is connected between the second port and a second end of the first element. Second ends of the at least two second elements are connected to each other and then connected to the resonator.

Abstract: The present disclosure relates to systems, methods, and non-transitory computer readable media that utilize machine learning to generate subject lines from subject line keywords. In one or more embodiments, the disclosed systems receive, from a client device, one or more subject line keywords. Additionally, the disclosed systems generate, utilizing a subject generation machine-learning model having learned parameters, a subject line by selecting one or more words for the subject line from a word distribution based on the one or more subject line keywords. The disclosed systems further provide, for display on the client device, the subject line.

Abstract: A secondary battery includes an electrode assembly which includes a positive electrode sheet, a negative electrode sheet, and a separator. The positive electrode sheet includes a positive electrode current collector and a positive electrode active material layer. The positive electrode active material layer includes a positive electrode active material. The negative electrode sheet includes a negative electrode current collector and a negative electrode active material layer. The negative electrode active material layer includes a negative electrode active material. The positive electrode current collector as a whole extends to form a plurality of positive electrode tabs, and the negative electrode current collector as a whole extends to form a plurality of negative electrode tabs. A coating weight of the positive electrode active material layer is Wz, a coating weight of the negative electrode active material layer is Wf, and Wz and Wf satisfy: 1.6Wf?Wz?2.2Wf, and 3.25 mg/cm2?Wf?5.84 mg/cm2.

Abstract: A secondary battery includes a positive electrode sheet, a negative electrode sheet, a separator, and an electrolyte, where the positive electrode sheet includes a positive electrode active material layer. The negative electrode sheet includes a negative electrode active material layer. A coating weight of the positive electrode active material layer is Wz, a coating weight of the negative electrode active material layer is Wf, and Wz and Wf satisfy: 1.6Wf?Wz?2.2Wf, and 3.25 mg/cm2?Wf?5.84 mg/cm2. The separator includes a separator substrate, a first adhesive coating, and a second adhesive coating, where the first adhesive coating and the second adhesive coating are respectively provided on two sides of the separator substrate. The electrolyte includes chain carboxylate. Based on a mass of the electrolyte, a mass percentage of the chain carboxylate is 6% to 56%.

Abstract: A secondary battery includes a positive electrode sheet including a positive electrode active material, and a negative electrode sheet including a negative electrode active material, where Dv99 of the positive electrode active material is 27 ?m to 33 ?m, and Dv99 of the negative electrode active material is 23 ?m to 28 ?m; a separator includes a separator substrate, a first adhesive coating and a second adhesive coating, where the first adhesive coating and the second adhesive coating are respectively provided on two sides of the separator substrate, the first adhesive coating includes a first polymer binder, an average particle size of the first polymer binder is 0.3 ?m to 3 ?m, the second adhesive coating includes a second polymer binder, and an average particle size of the second polymer binder is 10 ?m to 38 ?m; and an electrolyte includes an organic solvent, a lithium salt, and an additive.

Abstract: A secondary battery includes a positive electrode plate, a negative electrode plate, and a separator, where the positive electrode plate includes a positive electrode active material layer, the positive electrode active material layer including a positive electrode active material; and the negative electrode plate includes a negative electrode active material layer, the negative electrode active material layer including a negative electrode active material. A coating weight of the positive electrode active material layer is Wz, and a coating weight of the negative electrode active material layer is Wf, satisfying: 1.6Wf?Wz?2.2Wf, and 3.25 mg/cm2?Wf?5.84 mg/cm2. Dv99 of the positive electrode active material is 27 ?m to 33 ?m, and Dv99 of the negative electrode active material is 23 ?m to 28 ?m. The separator includes a separator substrate, a first adhesive coating, and a second adhesive coating.

Abstract: A secondary battery includes a positive electrode sheet, a negative electrode sheet, and an electrolyte. The positive electrode sheet includes a positive electrode current collector and a positive electrode active material layer. The negative electrode sheet includes a negative electrode current collector and a negative electrode active material layer. The positive electrode current collector integrally extends to form multiple positive electrode tabs. The negative electrode current collector integrally extends to form multiple negative electrode tabs. The positive electrode active material layer includes a positive electrode active material, and Dv99 of the positive electrode active material is 27 ?m to 33 ?m. The negative electrode sheet includes the negative electrode active material layer, the negative electrode active material layer includes a negative electrode active material, and Dv99 of the negative electrode active material is 23 ?m to 28 ?m.

Abstract: The present disclosure provides a model training method based on a graph neural network, and the method includes: acquiring a first training graph and a label corresponding to the first training graph, wherein the first training graph is a relationship graph corresponding to training multimedia content, or the first training graph is a relationship graph corresponding to a training item; obtaining a position feature, a structure feature, and a node feature of the first training graph based on the first training graph; fusing the position feature, the structure feature, and the node feature, to obtain a fused feature; and training a target model based on the GNN according to the fused feature and the label.

Abstract: One aspect of the present disclosure pertains to an integrated circuit (IC) structure. The IC structure includes a die and an integrated heat sink structure disposed over the die. In some embodiments, the integrated heat sink structure includes a first closed-loop microchannel structure adjacent to the die and a second closed-loop microchannel structure disposed over the first closed-loop microchannel structure. In an example, the second closed-loop microchannel structure is disposed further away from the die than the first closed-loop microchannel structure. In some implementations, a plurality of microchannels and a micromixer chamber collectively provide the first and second closed-loop microchannel structures.

Abstract: A journey map system is configured to employ a machine-learning model to assign category tags to historical journey maps. The journey map system then generates clusters based on similarities between these journey maps. Effectiveness of these clusters in performing an operation is determined by the journey map system using performance indicator data. The journey map system identifies representative journey maps based on the performance indicator data, the clusters, and the category tags. Finally, the journey map system generates insights using a language model implemented with machine learning.

Abstract: A video stream acquisition method, apparatus and system, a device and a medium are provided. The video stream acquisition method is applied to a user terminal and includes: sending a stream pulling information acquisition request to a media server, to make the media server determine target stream pulling information based on a generated domain name; acquiring the target stream pulling information, performing a domain name resolution on the target stream pulling information based on a preset domain name resolution method, and obtaining the IP address of the target intranet edge server; and based on the IP address of the target intranet edge server, sending a target stream pulling request to the target intranet edge server, and acquiring a target video stream returned by the target intranet edge server based on the target stream pulling request.