CosalPure: Learning Concept from Group Images for Robust Co-Saliency Detection
East China Normal University（华东师范大学）；New York University（纽约大学）；Nanyang Technological University（南洋理工大学）；Shanghai Industrial Control Safety Innovation Tech. Co., Ltd（上海工业控制安全创新科技有限公司）
Jiayi Zhu, Qing Guo, Felix Juefei-Xu, Yihao Huang, Yang Liu, Geguang Pu
Co-salient object detection (CoSOD) aims to identify the common and salient (usually in the foreground) regions across a given group of images. Although achieving significant progress, state-of-the-art CoSODs could be easily affected by some adversarial perturbations, leading to substantial accuracy reduction. The adversarial perturbations can mislead CoSODs but do not change the high-level semantic information (e.g., concept) of the co-salient objects. In this paper, we propose a novel robustness enhancement framework by first learning the concept of the co-salient objects based on the input group images and then leveraging this concept to purify adversarial perturbations, which are subsequently fed to CoSODs for robustness enhancement. Specifically, we propose CosalPure containing two modules, i.e., group-image concept learning and concept-guided diffusion purification. For the first module, we adopt a pre-trained text-to-image diffusion model to learn the concept of co-salient objects within group images where the learned concept is robust to adversarial examples. For the second module, we map the adversarial image to the latent space and then perform diffusion generation by embedding the learned concept into the noise prediction function as an extra condition. Our method can effectively alleviate the influence of the SOTA adversarial attack containing different adversarial patterns, including exposure and noise. The extensive results demonstrate that our method could enhance the robustness of CoSODs significantly.
Deep saliency: What is learnt by a deep network about saliency?
University of Exeter（埃克塞特大学）；Department of Computer Science, College of Engineering, Mathematics and Physical Sciences, University of Exeter, UK（埃克塞特大学工程、数学与物理科学学院计算机科学系（注：此处按要求提取到大学级别，故主要机构为University of Exeter））
Forget-It-All: Multi-Concept Machine Unlearning via Concept-Aware Neuron Masking
The University of Arizona（亚利桑那大学）；The University of Tulsa（塔尔萨大学）；Clemson University（克莱姆森大学）；Western Digital Corporation（西部数据公司）；University of Georgia（佐治亚大学）
SACE: Concept Erasure at the Semantic Singularity in Visual Autoregressive Models
School of Computer Science and Engineering, Sun Yat-sen University（中山大学计算机科学与工程学院）；School of Computer Science and Technology, Beijing Institute of Technology（北京理工大学计算机科学与技术学院）
Enhancing Concept Localization in CLIP-based Concept Bottleneck Models
Unité d’Informatique et d’Ingénierie des Systèmes ENSTA Paris, Institut Polytechnique de Paris（ENSTA巴黎信息与系统工程单位，巴黎理工学院）；Department of Information Engineering and Computer Science University of Trento（信息工程与计算机科学系，特伦托大学）
Envisioning Beyond the Few: Disentangled Semantics and Primitives for Few-Shot Atypical Layout-to-Image Generation
State Key Laboratory of Virtual Reality Technology and Systems（虚拟现实技术与系统国家重点实验室）；School of Computer Science and Engineering（计算机科学与工程学院）；Qingdao Research Institute, Beihang University, China（北京航空航天大学青岛研究所，中国）
East China Normal University（东华大学）；New York University（纽约大学）；IHPC & CFAR, Agency for Science, Technology and Research（IHPC与CFAR，科技研究局）；Nanyang Technological University（南洋理工大学）；Shanghai Industrial Control Safety Innovation Tech. Co., Ltd（上海工业控制安全创新技术有限公司）
UCSF（加州大学旧金山分校）；SFVAMC（旧金山退伍军人事务医疗中心）；UC Berkeley（加州大学伯克利分校）；Helen Diller Family Comprehensive Cancer Center, UCSF（加州大学旧金山分校海伦·迪勒家族综合癌症中心）；Bakar Computational Health Sciences Institute, UCSF（加州大学旧金山分校巴卡尔计算健康科学研究所）
Discovering Fine-Grained Visual-Concept Relations by Disentangled Optimal Transport Concept Bottleneck Models
National Key Laboratory of Radar Signal Processing, Xidian University, Xi’an, 710071, China（雷达信号处理国家级重点实验室，西安电子科技大学）；State Key Laboratory of Integrated Service Networks, Xidian University, Xi’an, 710071, China（集成服务网络国家重点实验室，西安电子科技大学）
Classification Metrics for Image Explanations: Towards Building Reliable XAI-Evaluations
Fraunhofer Institute for Manufacturing Engineering and Automation IPA（弗劳恩霍夫制造工程与自动化研究所）；Institute of Industrial Manufacturing and Management IFF, University of Stuttgart（斯图加特大学工业制造与管理研究所）
MACE: Model Agnostic Concept Extractor for Explaining Image Classification Networks
IIT Ropar（印度理工学院罗帕尔分校）
深度卷积网络在各种图像分类任务中非常成功。目前解释预训练模型预测的方法依赖于梯度信息，通常会导致突出图集中在整个前景对象上。然而，人类通常通过解剖图像并指出较小概念的存在来进行推理。最终的输出通常是这些较小概念的存在或不存在的集合。在这项工作中，我们提出了MACE：一个模型不可知的概念提取器，它可以通过更小的概念来解释卷积网络的工作。MACE框架剖析了由卷积网络为图像生成的特征图，以提取基于概念的原型解释。此外，它估计提取的概念与预训练模型预测的相关性，这是解释单个类预测所需的关键方面，在现有方法中缺失。我们使用VGG16和ResNet50 CNN架构以及Animals With Attributes 2（AWA2）和Places365等数据集验证了我们的框架。我们的实验表明，由MACE框架提取的概念增加了人类对解释的可解释性，并且忠实于底层的预训练黑盒模型。
Inspiring the Next Generation of Segment Anything Models: Comprehensively Evaluate SAM and SAM 2 with Diverse Prompts Towards Context-Dependent Concepts under Different Scenes
Yale University, USA（耶鲁大学）；Nanyang Technological University, Singapore（南洋理工大学）；Dalian University of Technology, China（大连理工大学）；X3000 Inspection Co., Ltd, China（X3000检测有限公司）
SAP-DETR: Bridging the Gap Between Salient Points and Queries-Based Transformer Detector for Fast Model Convergency
Institute of Computing Technology (ICT), Chinese Academy of Sciences（中国科学院计算技术研究所）；Stanford University（斯坦福大学）；University of Chinese Academy of Sciences（中国科学院大学）；AI Lab, Lenovo Research（联想研究院人工智能实验室）；Lenovo Ltd.（联想集团有限公司）
