4results about How to "Valid update" patented technology

This application relates to the field of semantic analysis technology, specifically to a method and system for extracting audit anomaly semantics from unstructured data. The method includes: acquiring a benchmark graph and a benchmark database and their entity mentions; obtaining the strings of entity mentions based on their positions; acquiring an audit candidate set of strings from the benchmark database; determining a matching score based on the consistency of audit roles, and then filtering the candidate entity set; encoding the candidate entities, their audit role sets, entity mentions, and strings to obtain the probability distribution of entity mentions and candidate entities, and determining the audit role consistency score based on the probability distribution; classifying the association between entity mentions and candidate entities based on the overall magnitude of the audit role consistency score, and applying different audit rules to each category; and extracting audit anomaly semantics using the audit rules. This application improves the accuracy and efficiency of semantic extraction.

The application relates to the technical field of micro-nano manufacturing and electrochemical additive manufacturing, and more particularly to a limited electrodeposition device, system and method based on two-phase temperature regulation. The device comprises a platform, a cathode substrate installed on the platform, a moving mechanism installed on the platform, a nozzle installed on the moving end of the moving mechanism, a heating assembly arranged on a liquid supply path of the nozzle and a cooling assembly installed on the platform; the heating assembly is used for heating working liquid to be higher than room temperature; a water flow channel is arranged in the nozzle; the surface of the cathode substrate is covered with a mask layer, and the cooling assembly is used for cooling the mask layer to be below room temperature; and the moving mechanism is used for driving the nozzle to move relative to the cathode substrate. Through the temperature gradient of the high-temperature working liquid and the low-temperature liquid mask, a thermal Marangoni convection is induced at the liquid-liquid interface, the convection can effectively update the concentration boundary layer, the high-concentration ions in the body are continuously sucked to the deposition interface, the mass transfer bottleneck at room temperature is broken, and the limiting current density and the deposition rate are greatly improved.

This invention discloses an efficient high-resolution image visual tag generation method for multimodal large-scale models. The invention includes the following steps: First, the input original image is segmented using a minimum-fill dynamic slicing method to obtain several image patches, which are then added to a set of image patches to be processed. The original image is preprocessed and also added to the set of image patches to be processed. Next, a visual encoder is used to extract multi-layer visual features from each image patch in the set of image patches to be processed. Then, the penultimate layer of features is downsampled to low-resolution features and cross-attention is performed with the high-resolution features to obtain a visual tag subsequence for the current image patch. Finally, the final compressed visual tag sequence is obtained. This invention can efficiently generate visual tag sequences and is applicable to various visual reasoning tasks such as visual question answering, document question answering, and optical character recognition, providing an efficient and accurate visual context representation method for multimodal large-scale language models.

The application discloses a kind of life prediction method of power semiconductor module, terminal equipment and storage medium, determine the service life influencing factor of power semiconductor device under the operation of power cycle condition;From the influence factor, construct training data set, and the data in training data set is normalized, the training data set after normalization is used to train deep convolutional neural network, and the life prediction model is obtained.The present application considers the individual and mutual influence of characteristic parameters on service life, and produces more accurate prediction results.