Large Language Models (LLMs) have shown underperformance in information extraction (IE) tasks compared to smaller models. This underperformance is largely attributed to the mismatch between IE's structured output ...
详细信息
Self-supervised learning usually uses a large amount of unlabeled data to pre-train an encoder which can be used as a general-purpose feature extractor, such that downstream users only need to perform fine-tuning oper...
Self-supervised learning usually uses a large amount of unlabeled data to pre-train an encoder which can be used as a general-purpose feature extractor, such that downstream users only need to perform fine-tuning operations to enjoy the benefit of "large model". Despite this promising prospect, the security of pre-trained encoder has not been thoroughly investigated yet, especially when the pre-trained encoder is publicly available for commercial *** this paper, we propose AdvEncoder, the first framework for generating downstream-agnostic universal adversarial examples based on the pre-trained encoder. AdvEncoder aims to construct a universal adversarial perturbation or patch for a set of natural images that can fool all the downstream tasks inheriting the victim pre-trained encoder. Unlike traditional adversarial example works, the pre-trained encoder only outputs feature vectors rather than classification labels. Therefore, we first exploit the high frequency component information of the image to guide the generation of adversarial examples. Then we design a generative attack framework to construct adversarial perturbations/patches by learning the distribution of the attack surrogate dataset to improve their attack success rates and transferability. Our results show that an attacker can successfully attack downstream tasks without knowing either the pre-training dataset or the downstream dataset. We also tailor four defenses for pre-trained encoders, the results of which further prove the attack ability of AdvEncoder. Our codes are available at: https://***/CGCL-codes/AdvEncoder.
Diffusion models have made significant contributions to computer vision, sparking a growing interest in the community recently regarding the application of them to graph generation. Existing discrete graph diffusion m...
详细信息
Diffusion models have made significant contributions to computer vision, sparking a growing interest in the community recently regarding the application of them to graph generation. Existing discrete graph diffusion models exhibit heightened computational complexity and diminished training efficiency. A preferable and natural way is to directly diffuse the graph within the latent space. However, due to the non-Euclidean structure of graphs is not isotropic in the latent space, the existing latent diffusion models effectively make it difficult to capture and preserve the topological information of graphs. To address the above challenges, we propose a novel geometrically latent diffusion framework HypDiff. Specifically, we first establish a geometrically latent space with interpretability measures based on hyperbolic geometry, to define anisotropic latent diffusion processes for graphs. Then, we propose a geometrically latent diffusion process that is constrained by both radial and angular geometric properties, thereby ensuring the preservation of the original topological properties in the generative graphs. Extensive experimental results demonstrate the superior effectiveness of HypDiff for graph generation with various topologies. Copyright 2024 by the author(s)
Deep learning has significantly advanced time series forecasting through its powerful capacity to capture sequence relationships. However, training these models with the Mean Square Error (MSE) loss often results in o...
详细信息
An emerging area of research aims to learn deep generative models with limited training data. Prior generative models like GANs and diffusion models require a lot of data to perform well, and their performance degrade...
computing the weighted girth, which is the sum of weights of edges in the minimum weight cycle,is an important problem in network analysis. The problem for distributively computing girth in unweighted graphs has garne...
详细信息
computing the weighted girth, which is the sum of weights of edges in the minimum weight cycle,is an important problem in network analysis. The problem for distributively computing girth in unweighted graphs has garnered lots of attention, but there are few studies in weighted graphs. In this paper, we propose a distributed randomized algorithm for computing the weighted girth in weighted graphs with integral edge weights in the range [1, nc], where n is the number of vertices and c is a constant. The algorithm is devised under the standard synchronous CON GE S T model, which limits each vertex can only transfer O（log n） bits information along each incident edge in a round. The upper bound of the algorithm is O（n log2n） rounds. We also prove the lower bound for computing the weighted girth is ?（D + n/log n） where D is the hop diameter of the weighted graph. This means our distributed algorithm is optimal within a factor of O（log3n）.
Spiking Neural Networks (SNNs), renowned for their low power consumption, brain-inspired architecture, and spatio-temporal representation capabilities, have garnered considerable attention in recent years. Similar to ...
详细信息
Retrieval-Augmented Large Language Models (RALMs) have made significant strides in enhancing the accuracy of generated responses. However, existing research often overlooks the data quality issues within retrieval res...
详细信息
Sponsored online advertising delivers many billions of revenues for online ads publishers. The ads systems take userinput query keywords and display ads that are relevant to the query. the task of click-through rate (...
详细信息
ISBN:
(纸本)9781665480468
Sponsored online advertising delivers many billions of revenues for online ads publishers. The ads systems take userinput query keywords and display ads that are relevant to the query. the task of click-through rate (CTR) prediction aims to estimate the likelihood of a user clicking on the ads, which has become one of the core goals in the ads system. In order to further improve the CTR, user portraits are also considered as an input to make personalized ads display and recommendations, in the current deep learning CTR training platform. The naive combination of user space (~ 10 9 ) and feature space (~ 10] 12 however, would yield a 10 21 dimensional space. It is not only infeasible to feed the 10 21 parameters into the embedding layer with any off-the-shelf storage, but also impractical to train the network in such massive-scale dimensional space. In this paper, we design a novel CTR prediction framework for ads systems to tackle the massive-scale user-feature combination challenge. Specifically, we introduce a feature fusion network to explicitly learn user-feature cross embedding in an end-to-end manner. To improve the efficiency, we prune the feature fusion networks to a practical number through a network importance ranking scheme. Extensive empirical experiments on Baidu’s ads data validate the effectiveness of the proposed feature fusion networks.
The adversarial examples (AEs) cause misjudgments and damage the robustness of the DNNs systems. Previous studies have defended against AEs by detecting, but it is challenging to ensure a stable and high performance o...
详细信息
The adversarial examples (AEs) cause misjudgments and damage the robustness of the DNNs systems. Previous studies have defended against AEs by detecting, but it is challenging to ensure a stable and high performance of detecting AEs, while with a poor false detection. To this end, an AEs detection method named image reconstruction differences (IRD) is proposed to enhance the robustness of DNNs. Firstly, we use an end-to-end Com-Rec network to reconstruct examples with feature compression to expand the distinguishing features. Secondly, propose an image reconstruction differences based on information-theoretic VIF, structural information UQI and spectral information RASE composition to discriminate AEs. Moreover, we introduce the idea of integrated learning to form a strong random forest binary classifier to enhance the performance of detecting AEs. We further validate it through extensive experiments on the MNIST and CIFAR-10 datasets. These experiments demonstrated that the IRD effectively detected AEs and achieved a high average accuracy of 98.33%. Specifically it also performs favorably against the following methods based on Feature Squeezing, Local Intrinsic Dimensionality, Kernel Density and Network Invariance Checking with an average detection rate of 99.54% and a 1.44% average false positive rate.
暂无评论