Current popular Large Vision-Language Models (LVLMs) are suffering from Hallucinations on Object Attributes (HoOA), leading to incorrect determination of fine-grained attributes in the input images. Leveraging significant advancements 3D generation a single image, this paper proposes novel method mitigate HoOA LVLMs. This utilizes multiview images sampled generated representations as visual prompts for LVLMs, thereby providing more information other viewpoints. Furthermore, we observe order...

Meta-learning aims to teach the machine how learn. Embedding model-based meta-learning performs well in solving few-shot problem. The methods use an embedding model, usually a convolutional neural network, extract features from samples and classifier measure extracted particular stage of model. However, feature model at low contains richer visual information, while high semantic information. Existing fail consider impact information carried by different stages on performance classifier....

Irregularly sampled time series are ubiquitous, presenting significant challenges for analysis due to missing values. Despite existing methods address imputation, they predominantly focus on leveraging intra-series information, neglecting the potential benefits that inter-series information could provide, such as reducing uncertainty and memorization effect. To bridge this gap, we propose PRIME, a Prototype Recurrent Imputation ModEl, which integrates both imputing values in irregularly...

Accurate recognition of tool state is important for maximizing life. However, the sensor data collected in real-life scenarios has unbalanced characteristics. Additionally, although graph neural networks (GNNs) show excellent performance feature extraction spatial dimension data, it difficult to extract features temporal efficiently. Therefore, we propose a method based on Pruned Optimized Graph Neural Network-Gated Recurrent Unit (POGNN-GRU) under data. Firstly, design Improved-Majority...

Abstract Compared with three-phase motors, multi-phase motor speed control systems have many advantages, which make them good prospects in fields, such as the system of electric vehicle power equipment. In these systems, it is necessary for to a wider range. This paper takes six-phase permanent magnet synchronous motors research object, and analyzes several methods vector theory based on derivation its mathematical model, including i d = 0 control, maximum torque per ampere (MTPA) flux...

Space Environment Simulation and Research Infrastructure (SESRI) is one of the important technologies to promote development aerospace industry, ion accelerator system more subsystems in SESRI. Aiming at problem that temperature too high stability electric field affected by long-term operation accelerator, this paper designs an monitoring based on EPICS. In paper, control principle firstly analyzed, according its characteristics, robust realized using PID algorithm. Secondly, EPICS, overall...

The aim of meta-learning is to train the machine learn quickly and accurately. Improving performance model important in solving problem small samples achieving general artificial intelligence. A method based on feature embedding that exhibits good few-shot was previously proposed. In this method, pretrained deep convolution neural network used as sample features, output one layer representation samples. main limitation inability fuse low-level texture features high-level semantic joint...

In the field of information and communication, data model is used to describe organization form information. It pilot work establish corresponding database. Space Environment Simulation Research Infrastructure, various. This article first analyzes needs be collected a unified for monitoring based on Infrastructure established; Secondly, adaptability was analyzed conversion plan non-unified designed; Finally, UaModeler constructed OPC UA information, communication protocol efficiency...

This paper describe the principles of Servo Control System Electrical Discharge Machining Based on PMAC .To meet requirements processing and Put forward based servo control system. system combines hardware resource with real-time function. Elaborate human-computer interface developing process.

Micro-EDM is a flexible process which has many technique methods such as forming, drilling, milling, copying and WEDM. Combined these together can enhance the EDM machining ability. In this paper set of route design regulations used in combined proposed. Through regulations, optimal be obtained improve high quality efficiency. To test performance micro part designed its fabrication performed at last.

Abstract In large-scale complex knowledge graphs (KGs) for reasoning, most existing methods ignore the hierarchical features of KGs, limiting their efficiency in recommender systems, information retrieval, and intelligent Q&A systems by discrete nature. Based on translation-based graph embedding method, this paper proposes a method which encodes semantics entity representations through continuous bag-of-word (CBOW) convolutional neural network (CNN) models. At same time, hierarchies are...

This paper aims to investigate the performance of narrow beam machined by MWEDM. The machining experiments are carried out on self-development Micro-WEDM with Φ30μm electrode wire. Narrow beams width around 10μm E52100 steel under different machine conditions (discharge energy, servo reference voltage). Based analysis results, coupled temperature and stress using ANSYS finite element software is performed causes failures process.

In Micro Wire Electrical Discharge Machining (Micro-WEDM), the kerf width varies with different machining parameters, which will greatly influence precision. order to study variations in Micro-WEDM, of variation is analyzed and experiment considering speed are performed on self-developed micro-WEDM under parameters. With reference results, 32μm wide slot machined Φ30μm wire-tool stainless steel workpiece.

The Space Environment Simulation and Research Infrastructure (SESRI) is one of the key technologies driving development space industry. micromechanical analysis subsystem an important part SESRI. Helium plays role in start-up, operation, testing some equipment subsystem. Once a leak occurs, it often undetectable to experimenters. In order solve problem helium leaks, this paper designs real-time monitoring alarm system based on Phoebus/Alarms. uses analog acquisition module Siemens CPU S7-300...

Compared with open experimental equipment, closed equipment that uses computer software a private protocol to complete the process has higher security and is widely used in fields such as space environment experiment simulation. This paper takes of research object, while proposes cooperative monitoring strategy for process. By combining image structure similarity comparison algorithm dynamic event monitoring, this method realizes proprietary software. The test results show proposed better...