薄荷奶绿Yena-CSDN博客

原创【TAROT学习日记】韦特体系塔罗牌学习（5）——皇帝 THE EMPEROR IV

他有点武断的倾向，但是当他在许下承诺后，通常是值得信赖的。而当你也允下承诺时，他会期望你能实行它，假如你做不到，他是不会再给你第二次机会的。一国之尊的皇帝充满了自信，头戴皇冠，身着红袍，足蹬象征严格纪律的战靴，自信满满的端坐在宝座上。皇帝牌代表道德戒律与禁令，此牌偏向次序，规范和稳定的局面，拥有至高无上的的权利。这是描述有男性内在特质的人事物，这里并不一定指的是人，也可以形容在处理问题上的风格。：组织的混乱，事业的挫败；

2024-07-19 22:25:42 642

原创【TAROT学习日记】韦特体系塔罗牌学习（3）——女祭司 THE HIGH PRIESTESS II

女祭司端坐于石凳之上，手上的卷宗上书写着 Tora，乃律法之一，代表她内心的知性和洞察力，长短分明，善恶明暗，也代表神秘的魔法，静静等待适合行动的时机，与给于明确的指示。处在对立矛盾之中，表面静默但是内心波涛汹涌，极具洞察力，感悟人性沧桑。塔罗牌女祭司相传是月亮的使者，虔诚，静默却又透着十分的警觉和睿智，是智慧与直觉的象征。：有可能从事需要深度思考的工作，倾向于研究和学习；感情生活混乱，不能合适的处理；柏拉图式的恋情，精神交流；：未来发展还不明朗，但维持现状的可能性较高，保守看待。

2024-07-16 21:37:00 649

原创【TAROT学习日记】韦特体系塔罗牌学习（2）——魔术师 THE MAGICIAN I

塔罗牌魔术师正在实行法术，他右手高举权杖指向天，召唤神奇的法力，左手食指指向地，象征了他是联合天与地的桥梁，主宰者地上的一切。魔术师在塔罗牌中的编号是1，预示着一切的开始，同时也扮演者大阿尔卡纳与小阿尔卡之间地沟通。他不是被生活的潮流推着走，而是在这些潮流中为他自己行动，并做出具体的成果。有一种适合自己的学习方式，并取得好成绩；：面对一个新的工作，把握机会，发挥自己的能力，会得到想要的东西。：一切还在起步的阶段，但是仍然有许多发展的空间，按部就班完成。对的时间，对的机会，对的动机，使你的努力获得好的结果。

2024-07-15 20:13:15 899

原创【TAROT学习日记】韦特体系塔罗牌学习（1）——愚者 THE FOOL 0

活在过去和未来的人可能认为他对眼前事物的执着是愚蠢的，但他并不在意别人的看法。：工作方面，喜欢寻求捷径，倾向于自由的工作氛围，适合艺术类工作或从事自由职业。学业方面，出于好奇心对当前的学业产生浓厚的兴趣，善于把握重点，容易以独特的方式取自意外的收获。这是一个新的开始，你正在开始一次冒险，你有无限的可能性，你有无限的潜力，你是天真的，你是乐观的。：你正在面对很危险的状况，最重要的是考虑如何明哲保身，不要玩火自焚。新的起点，不能再等了，自由，好奇，直觉，纯真；：很不稳定的工作，跳出去会有更多的好机会。

2024-07-11 02:00:00 667

原创【扩散对抗】AdvDiffuser: Natural Adversarial Example Synthesis with Diffusion Models

原文标题： AdvDiffuser: Natural Adversarial Example Synthesis with Diffusion Models原文代码： https://github.com/lafeat/advdiffuser发布年度： 2023发布期刊： ICCVPrevious work on adversarial examples typically involves a fixed norm perturbation budget, which fails to captu

2024-07-10 15:05:23 593

原创【VQA】Consistency and Uncertainty: Identifying Unreliable Responses From Black-Box Vision-Language Mo

原文标题： Consistency and Uncertainty: Identifying Unreliable Responses From Black-Box Vision-Language Models for Selective Visual Question Answering原文代码：暂无发布年度： 2024发布期刊： CVPRThe goal of selective prediction is to allow an a model to abstain when it may n

2024-07-10 15:04:40 675

原创机器学习的遗忘——基于文章“Forgetting“ in Machine Learning and Beyond: A Survey

另一个需要考虑的重要方面是偶然遗忘的界限，因为并非所有的遗忘情况都会产生积极的结果。在教育理论中，过度遗忘可能会使问题解决变得过于具有挑战性，潜在地阻碍学生的学习流程通道，从而产生负面体验同样，不适当的存档实践可能会导致机器学习模型的灾难性遗忘和荒谬的决策制定，这给我们带来了新的挑战，即确定不同类型训练任务中适当的遗忘界限，确保遗忘被有效和适当地应用。这种形式的遗忘有助于优化模型的内存利用，提高其概括能力，并增强其对不同数据集和任务的适应性。在这里，我们探索当前的研究空白，以推动该领域的发展。

2024-07-08 11:14:52 541

原创【对抗去偏】BiasAdv: Bias-Adversarial Augmentation for Model Debiasing

原文标题： BiasAdv: Bias-Adversarial Augmentation for Model Debiasing原文代码：暂无发布年度： 2023发布期刊： CVPRNeural networks are often prone to bias toward spurious correlations inherent in a dataset, thus failing to generalize unbiased test criteria. A key challenge to

2024-06-21 20:37:41 747

原创【多模态攻击】Set-level Guidance Attack: Boosting Adversarial Transferability of Vision-Language Pre-train

原文标题： Set-level Guidance Attack: Boosting Adversarial Transferability of Vision-Language Pre-training Models原文代码： https://github.com/Zoky-2020/SGA发布年度： 2023发布期刊： ICCVVision-language pre-training (VLP) models have shown vulnerability to adversarial examp

2024-06-21 20:35:22 908

原创【图像攻击转移性】FACL-Attack: Frequency-Aware Contrastive Learning for Transferable Adversarial Attacks

原文标题： FACL-Attack: Frequency-Aware Contrastive Learning for Transferable Adversarial Attacks原文代码：暂无发布年度： 2024发布期刊： AAAIDeep neural networks are known to be vulnerable to security risks due to the inherent transferable nature of adversarial examples. De

2024-05-27 16:28:41 884

原创【多模态对抗】VQAttack: Transferable Adversarial Attacks on Visual Question Answering via Pre-trained Model

原文标题： VQAttack: Transferable Adversarial Attacks on Visual Question Answering via Pre-trained Models原文代码： https://github.com/ericyinyzy/VQAttack发布年度： 2024发布期刊： AAAIVisual Question Answering (VQA) is a fundamental task in computer vision and natural lang

2024-05-27 16:27:33 697

原创【对抗攻击转移性】Boosting Adversarial Transferability by Block Shuffle and Rotation

原文标题： Boosting Adversarial Transferability by Block Shuffle and Rotation原文代码： https://github.com/Trustworthy-AI-Group/BSR发布年度： CVPR发布期刊： 2024Adversarial examples mislead deep neural networks with imperceptible perturbations and have brought significant

2024-04-01 20:30:33 892

原创【图像对抗攻击】Towards Transferable Adversarial Attacks with Centralized Perturbation

原文标题： Towards Transferable Adversarial Attacks with Centralized Perturbation原文代码：暂无发布年度： 2024发布期刊： AAAIAdversarial transferability enables black-box attacks on unknown victim deep neural networks (DNNs), rendering attacks viable in real-world scenarios

2024-04-01 20:30:08 789

原创 [物理对抗攻击]Adversarial Attack with Raindrops

原文标题： Adversarial Attack with Raindrops原文代码：暂无发布年度： 2023发布期刊： CVPRDeep neural networks (DNNs) are known to be vulnerable to adversarial examples, which are usually designed artificially to fool DNNs, but rarely exist in real-world scenarios. In this pa

2024-03-29 15:26:36 870 1

原创【多模态对抗攻击】VLATTACK: Multimodal Adversarial Attacks on Vision-Language Tasks via Pre-trained Models

原文标题： VLATTACK: Multimodal Adversarial Attacks on Vision-Language Tasks via Pre-trained Models原文代码： https://github.com/ericyinyzy/VLAttack发布年度： 2023发布期刊： NeurIPSVision-Language (VL) pre-trained models have shown their superiority on many multimodal task

2024-03-29 15:25:55 1158

原创【多模态对抗】AdvCLIP: Downstream-agnostic Adversarial Examples in Multimodal Contrastive Learning

原文标题： AdvCLIP: Downstream-agnostic Adversarial Examples in Multimodal Contrastive Learning原文代码： https://github.com/CGCL-codes/AdvCLIP发布年度： 2023发布期刊： ACM MMMultimodal contrastive learning aims to train a general-purpose feature extractor, such as CLIP, o

2024-03-07 16:03:29 978 1

原创【文本对抗攻击】Bridge the Gap Between CV and NLP!A Gradient-based Textual Adversarial Attack Framework

原文标题： Bridge the Gap Between CV and NLP!A Gradient-based Textual Adversarial Attack Framework原文代码： https://github.com/Phantivia/T-PGD发布年度： 2023发布期刊： ACLDespite recent success on various tasks, deep learning techniques still perform poorly on adversari

2024-03-07 16:01:48 838 1

原创【对抗样本】Downstream-agnostic Adversarial Examples

原文标题： Downstream-agnostic Adversarial Examples原文代码： https://github. com/CGCL-codes/AdvEncoder发布年度： 2023发布期刊： ICCVSelf-supervised learning usually uses a large amount of unlabeled data to pre-train an encoder which can be used as a general-purpose featur

2024-02-29 22:18:06 990

原创【语义扰动的对抗攻击】Mutual-modality Adversarial Attack with Semantic Perturbation

原文标题： Mutual-modality Adversarial Attack with Semantic Perturbation原文代码：暂无发布年度： 2024发布期刊： AAAIAdversarial attacks constitute a notable threat to machine learning systems, given their potential to induce erroneous predictions and classifications. Howeve

2024-02-29 22:17:37 890

原创【图检索】Context-I2W:Mapping Images to Context-dependent Words for Zero-Shot Composed Image Retrieval

原文标题： Context-I2W: Mapping Images to Context-dependent Words for Accurate Zero-Shot Composed Image Retrieval原文代码： https://github.com/Pter61/context i2w.发布年度： 2024发布期刊： AAAIDifferent from Composed Image Retrieval task that requires expensive labels for

2023-12-22 16:03:29 1035

原创【鲁棒VQA】Digging out Discrimination Information from Generated Samples for Robust VQA

原文标题： Digging out Discrimination Information from Generated Samples for Robust Visual Question Answering原文代码： https://github.com/Zhiquan-Wen/DDG发布年度： 2023发布期刊： ACLVisual Question Answering (VQA) aims to answer a textual question based on a given image.

2023-12-20 10:43:09 910 1

原创【图对抗】Local-Global Defense against Unsupervised Adversarial Attacks on Graphs

原文标题： Local-Global Defense against Unsupervised Adversarial Attacks on Graphs原文代码： https://github.com/jindi-tju/ULGD/blob/main发布年度： 2023发布期刊： AAAIUnsupervised pre-training algorithms for graph representation learning are vulnerable to adversarial attac

2023-12-18 00:01:57 820 1

原创【多模态攻击】Data Poisoning Attacks Against Multimodal Encoders

原文标题： Data Poisoning Attacks Against Multimodal Encoders原文代码： https://github.com/zqypku/mm_poison/发布年度： 2023发布期刊： ICMLRecently, the newly emerged multimodal models, which leverage both visual and linguistic modalities to train powerful encoders, have ga

2023-12-15 15:01:50 1040

原创【对抗vqa】Attacking VQA Systems via Adversarial Background Noise

原文标题： Attacking VQA Systems via Adversarial Background Noise原文代码： https://github.com/akshay107/vqa-adv-background发布年度： 2020发布期刊： IEEE TETCIAdversarial examples have been successfully generated for various image classification models. Recently, several m

2023-12-12 09:31:30 876

nbwjszd的博客