Large language models (LLMs) are widely applied in various natural language processing tasks such as question answering and machine translation. However, due to the lack of labeled data and the difficulty of manual an...
详细信息
In order to enter the era of utility, noisy intermediate-scale quantum (NISQ) devices need to enable long-range entanglement of large qubit chains. However, due to the limited connectivity of superconducting NISQ devi...
详细信息
ISBN:
(数字)9798350361759
ISBN:
(纸本)9798350361766
In order to enter the era of utility, noisy intermediate-scale quantum (NISQ) devices need to enable long-range entanglement of large qubit chains. However, due to the limited connectivity of superconducting NISQ devices, long-range entangling gates are realized in linear depth. Furthermore, a time-dependent degradation of the average CNOT gate fidelity is observed. Likely due to aging, this phenomenon further degrades entanglement capabilities. Our aim is to help in the current efforts to achieve utility and provide an opportunity to extend the utility lifespan of current devices -albeit by selecting fewer, high quality resources. To achieve this, we provide a method to transform user-provided CNOT and readout error requirements into a compliant partition onto which circuits can be executed. We demonstrate an improvement of up to 52 % in fidelity for a random CNOT chain of length 50 qubits and consistent improvements between 11.8% and 47.7% for chains between 10 and 40 in varying in increments of 10, respectively.
This paper investigates the role of CLIP image embeddings within the Stable Video Diffusion (SVD) framework, focusing on their impact on video generation quality and computational efficiency. Our findings indicate tha...
详细信息
Under the traditional mode of classroom teaching, there is a problem that students' classroom behaviours are inconsistent with their classroom results. In view of the above problems, this paper analyses students&#...
详细信息
In this paper, the method to design deep Convolutional Neural Network (CNN) architecture for the problem of traffic signs classification is proposed. The approach incorporates five main stages followed by each other: ...
详细信息
At present, information technology is widely used in various industries, and the application area of information technology is constantly expanding. Information technology is an important part of modern society, and i...
详细信息
Standards such as ISO 13849 and ISO 12100 enable users to model safety related control elements with safety functions, according to a specified architecture and required performance level. In this direction, a novel A...
详细信息
We develop the Bayesian Wasserstein repulsive Gaussian mixture model that promotes well-separated clusters. Unlike existing repulsive mixture approaches that focus on separating the component means, our method encoura...
Unsupervised Domain Adaptation Semantic Segmentation (UDASS) aims to harness labeled data from the source domain alongside unlabeled data from the target domain to effectively segment target domain images. While self-...
详细信息
ISBN:
(数字)9798350359312
ISBN:
(纸本)9798350359329
Unsupervised Domain Adaptation Semantic Segmentation (UDASS) aims to harness labeled data from the source domain alongside unlabeled data from the target domain to effectively segment target domain images. While self-training has achieved tremendous success in UDASS, most existing methods heavily rely on source data and lack of sufficient learning from target data, resulting in a performance decline. To address this issue, this paper introduces a Bidirectional Regularization Guidance (BRG) method, which combines a Target Perturbation Consistency (TPC) module and a thing-class ImageNet Feature Distance Reweighting (FDR) module to provide effective regularization guidance. Specifically, the TPC module is introduced to perturb the target stream by masking at the input level and adding feature noise at the feature level. This module employs a target perturbation consistency loss to penalize inconsistencies between perturbed student predictions and teacher model predictions, thereby facilitating improved learning of contextual information by the student model from the target images. To further enhance the model’s generalization ability to the target domain, We propose an FDR module that employs a transferability map to calculate a reweighted graph, helping to effectively align source features with ImageNet features by reweighting the feature distances, especially those with lower transferability Through rigorous experimentation in standard UDASS settings—training with synthetic labeled and real unlabeled data—BRG outperforms baseline model on GTAV→Cityscapes and SYNTHIA→Cityscapes.
Domain-specific modeling languages abstractly represent domain knowledge in a way that users can more easily understand the model content without technical expertise. These languages can be created for any domain, pro...
详细信息
暂无评论