In this paper,we introduce the large language model and domain-specific model collaboration(LDMC)framework designed to enhance smart *** LDMC framework leverages the comprehensive and versatile knowledge of large doma...
详细信息
In this paper,we introduce the large language model and domain-specific model collaboration(LDMC)framework designed to enhance smart *** LDMC framework leverages the comprehensive and versatile knowledge of large domain-general models,combines it with the specialized and disciplinary knowledge from small domainspecific models(DSMs),and incorporates pedagogy knowledge from learning theory *** integration yields multiple knowledge representations,fostering personalized and adaptive educational *** explore various applications of the LDMC framework in the context of smart education.
Federated learning (FL) is an emerging paradigm that allows participants to collaboratively train deep learning tasks while protecting the privacy of their local data. However, the absence of central server control in...
详细信息
Federated learning (FL) is an emerging paradigm that allows participants to collaboratively train deep learning tasks while protecting the privacy of their local data. However, the absence of central server control in distributed environments exposes a vulnerability to data poisoning attacks, where adversaries manipulate the behavior of compromised clients by poisoning local data. In particular, data poisoning attacks against FL can have a drastic impact when the participant's local data is non-independent and identically distributed (non-IID). Most existing defense strategies have demonstrated promising results in mitigating FL poisoning attacks, however, fail to maintain their effectiveness with non-IID data. In this work, we propose an effective defense framework, FL data augmentation (FLDA), which defends against data poisoning attacks through local data mixup on the clients. In addition, to mitigate the non-IID effect by exploiting the limited local data, we propose a gradient detection strategy to reduce the proportion of malicious clients and raise benign clients. Experimental results on datasets show that FLDA can effectively reduce the poisoning success rate and improve the global model training accuracy under poisoning attacks for non-IID data. Furthermore, FLDA can increase the FL accuracy by more than 12% after detecting malicious clients.
The human can easily recognize the incongruous parts of an image, for example, perturbations unrelated to the image itself, but are poor at spotting the small geometric transformations. However, in terms of the robust...
详细信息
The human can easily recognize the incongruous parts of an image, for example, perturbations unrelated to the image itself, but are poor at spotting the small geometric transformations. However, in terms of the robustness of deep neural networks (DNNs), the ability to properly recognize objects with small geometric transformations is still a challenge. In this work, we investigate the problem from the perspective of adversarial attacks: does the performance of DNNs degrade even when small geometric transformations are applied to images? To this end, we propose a novel adversarial attack method, called WBA, a Warping-Based Adversarial attack method, which does not introduce information independent of the original images but manipulates the existing pixels of the images by elastic warping transformations to generate adversarial examples that are imperceptible to the human eye. At the same time, existing adversarial attacks typically generate adversarial examples by modifying pixels in the spatial domain of the image, the addition of such perturbations introduces extra information unrelated to the image itself and is easily detected by the naked eyes. We demonstrate the effectiveness of WBA by extensive experiments on commonly used datasets, including MNIST, CIFAR10, and ImageNet. The results show that WBA can quickly generate adversarial examples with the highest adversarial strength, consumes less time, and can be comparable to optimization-based adversarial attack methods in image perception evaluation metrics such as LPIPS, SSIM, and far more than gradient direction-based iterative methods.
The studies on superconductors under extreme conditions offer valuable insights for assessing their potential in new ***_(3)Sn,an intermetallic alloy with an A15 structure,is a key commercial superconductor known for ...
详细信息
The studies on superconductors under extreme conditions offer valuable insights for assessing their potential in new ***_(3)Sn,an intermetallic alloy with an A15 structure,is a key commercial superconductor known for its high critical current and magnetic field ***,we systematically investigated the physical properties of Nb_(3)Sn under high *** findings reveal that superconductivity in Nb_(3)Sn remains robust up to∼142 GPa,demonstrating remarkable stability despite a gradual suppression of��c with increasing ***-principles calculations indicate that the pressure-dependent superconducting behavior is primarily driven by variations in the density of states of Nb’s d-electrons,particularly contributions from the d_(x^(2)-y^(2)) and d_(z^(2)) ***,we predict the potential for synthesizing Nb_(3)Sn films and demonstrate that biaxial strain induced by suitable substrates can preserve their superconducting *** comprehensive study not only enhances our understanding of Nb_(3)Sn’s superconducting mechanism under high pressure but also opens new avenues for its application in advanced superconducting technologies.
To control the diffusion of high concentrations of coal dust during tunnel boring and minimize the threat to the life and health of coal miners, theoretical analysis, numerical simulations, and field measurements were...
详细信息
The growing memory demands of embedding tables in Deep Learning Recommendation Models (DLRMs) pose great challenges for model training and deployment. Existing embedding compression solutions cannot simultaneously ach...
详细信息
Multi-focus image fusion is a hot topic in the field of image processing, and it is a fundamental problem in the fields of image editing, image synthesis, and target retrieval. In previous fusion methods, although fea...
详细信息
Given a graph G and a query node q, community search (CS) seeks a cohesive subgraph from G that contains q. CS has gained much research interests recently. In the database research community, researchers aim to find t...
详细信息
The rapid development of DNA synthesis and sequencing technologies is making the ultra-high-density storage medium DNA to meet the rising demand for enormous data storage. The block storage interface, which is massive...
ISBN:
(纸本)9781939133458
The rapid development of DNA synthesis and sequencing technologies is making the ultra-high-density storage medium DNA to meet the rising demand for enormous data storage. The block storage interface, which is massively employed in storage systems, is the critical abstraction to integrate DNA storage into silicon-based computer systems. In this paper, we explore building block devices on DNA and identify the challenges of petabyte-scale metadata management and high DNA access costs. We propose a holistic DNA block device design called LIQUID-STATE DRIVE to provide low-cost block access to exabyte-scale data with the help of small yet fast SSDs. We adopt the dual-layer translation table to leverage SSDs to decrease the metadata updating cost. We introduce symbiotic metadata and delayed invalidation to reduce the cost of garbage collection and block updating. Our evaluation demonstrates that in microbenchmarks and real-world traces, the write cost reduces up to seven orders of magnitude and 2,927×, and the read cost reduces up to 6,206× and 7×, respectively. We expect our exploration and experience in building DNA block devices to be useful in expediting the advancement of DNA storage and bridging the gap between information technology and biotechnology.
The van der Waals heterojunctions,stacking of different two-dimensional materials,have opened unprecedented opportunities to explore new physics and device ***,combining the density functional theory with non-equilibr...
详细信息
The van der Waals heterojunctions,stacking of different two-dimensional materials,have opened unprecedented opportunities to explore new physics and device ***,combining the density functional theory with non-equilibrium Green’s function technique,we systematically investigate the spin-polarized transport properties of van der Waals magnetic tunnel junctions(MTJs),Cu/MnBi_(2)Te_(4)/MnBi_(2)Te_(4)/Cu and Cu/MnBi_(2)Te_(4)/hBN/n·MnBi_(2)Te_(4)/Cu(n=1,2,3).It is found that the maximum tunnel magnetoresistance of Cu/MnBi_(2)Te_(4)/hBN/3·MnBi_(2)Te_(4)/Cu MTJs can reach 162.6%,exceeding the system with only a single layer MnBi_(2)Te_(4).More interestingly,our results indicate that Cu/MnBi_(2)Te_(4)/h-BN/n·MnBi_(2)Te_(4)/Cu(n=2,3)MTJs can realize the switching function,while Cu/MnBi_(2)Te_(4)/h-BN/3·MnBi_(2)Te_(4)/Cu MTJs exhibit the negative differential *** Cu/MnBi_(2)Te_(4)/h-BN/3·MnBi_(2)Te_(4)/Cu in the parallel state shows a spin injection efficiency of more than 83.3%.Our theoretical findings of the transport properties will shed light on the possible experimental studies of MnBi_(2)Te_(4)-based van der Waals magnetic tunneling junctions.
暂无评论